lithofacies and petrophysical properties of mesaverde tight-gas

Lithofacies and Petrophysical Properties of Mesaverde Tight-Gas Sandstones in Western U.S. Basins: a short course

Alan P. Byrnes formerly Kansas Geological Survey- now Chesapeake Energy Robert M. Cluff John C. Webb Daniel A. Krygowski Stefani D. Whittaker The Discovery Group, Inc

2009 AAPG Annual Convention Short course #1 6 June 2009, Denver, Colorado

Cluff: Introduction and Overview

Lithofacies and Petrophysical Lithofacies and Petrophysical Properties of Mesaverde TightProperties of Mesaverde Tight--Gas Gas Sandstones in Western U.S. Basins: Sandstones in Western U.S. Basins: a short coursea short courseAlan P. Byrnes Alan P. Byrnes

formerlyformerly Kansas Geological SurveyKansas Geological Survey--now Chesapeake Energynow Chesapeake Energy

Robert M. Cluff Robert M. Cluff John C. Webb John C. Webb Daniel A. Krygowski Daniel A. Krygowski Stefani D. Whittaker Stefani D. Whittaker

The Discovery Group, IncThe Discovery Group, Inc

Denver, ColoradoAAPG ACE 2009: Denver Colorado 11

2009 AAPG Annual Convention2009 AAPG Annual ConventionShort course #1Short course #16 June 2009, Denver, Colorado6 June 2009, Denver, Colorado

Short course agendaShort course agenda8:008:00--8:308:30 Project overview, Bob CluffProject overview, Bob Cluff8:308:30--10:0010:00 Lithofacies and geology of the Lithofacies and geology of the

Mesaverde Group, John WebbMesaverde Group, John Webb10 0010 00 10 1510 15 b eakb eak10:0010:00--10:1510:15 breakbreak10:1510:15--noonnoon Porosity & permeability of Mesaverde Porosity & permeability of Mesaverde

tight gas sands, Alan Byrnestight gas sands, Alan Byrnesnoonnoon--1:00p1:00p lunchlunch1:001:00--2:302:30 Pc, resistivity, and relative Pc, resistivity, and relative

perm of Mesaverde, Alan Byrnesperm of Mesaverde, Alan Byrnes2:302:30--2:452:45 breakbreak2:452:45--4:154:15 Log evaluation of the Mesaverde Dan Log evaluation of the Mesaverde Dan

AAPG ACE 2009: Denver Colorado 2

2:452:45--4:154:15 Log evaluation of the Mesaverde, Dan Log evaluation of the Mesaverde, Dan Krygowski, Stefani Whittaker, Krygowski, Stefani Whittaker, & Bob Cluff& Bob Cluff

4:154:15--4:304:30 discussion, Q&A perioddiscussion, Q&A period

AAPG ACE Short Course 1: 06.06.2009 1 of 217


Project title:Analysis of Critical Permeability, Capillary and Electrical Properties for Mesaverde Tight Gas Sandstones

US DOE # DE-FC26-05NT42660US DOE # DE-FC26-05NT42660from Western U.S. Basins


website: http://www.kgs.ku.edu/mesaverdewebsite: http://www.kgs.ku.edu/mesaverde

Project overviewProject overviewProject proposal submitted on 21 March 2005 in Project proposal submitted on 21 March 2005 in response to DOE solicitation DEresponse to DOE solicitation DE--PS26PS26--04NT4272004NT42720DOE award DEDOE award DE--FC26FC26--05NT42660 in October 2005 05NT42660 in October 2005

for $411K DOE funds/$103K industry cofor $411K DOE funds/$103K industry co--shareshareDiscovery Group inDiscovery Group in--kind contribution of manpower and kind contribution of manpower and facilitiesfacilities

2 ½ year study with no2 ½ year study with no--cost extensioncost extensionAlan P. Byrnes, Principal InvestigatorAlan P. Byrnes, Principal InvestigatorUniversity of Kansas Center for Research was theUniversity of Kansas Center for Research was the


University of Kansas Center for Research was the University of Kansas Center for Research was the umbrella contracting organizationumbrella contracting organization

Kansas Geological Survey and The Discovery Group, coKansas Geological Survey and The Discovery Group, co--participating research contractorsparticipating research contractors



Team MembersTeam Members

University of KansasUniversity of Kansas--Kansas Geological SurveyKansas Geological SurveyAlan P. Byrnes (Principal Investigator)Alan P. Byrnes (Principal Investigator)Support Team Members: Support Team Members: John Victorine, Ken Stalder, Daniel S. Osburn, John Victorine, Ken Stalder, Daniel S. Osburn, Andrew Knoderer, Owen Metheny, Troy Andrew Knoderer, Owen Metheny, Troy Hommertzheim, Joshua P. ByrnesHommertzheim, Joshua P. Byrnes

The Discovery Group, Inc.The Discovery Group, Inc.


The Discovery Group, Inc.The Discovery Group, Inc.Robert M. Cluff (coRobert M. Cluff (co--Principal Investigator)Principal Investigator)John C. Webb, Daniel A. Krygowski, Stefani WhittakerJohn C. Webb, Daniel A. Krygowski, Stefani Whittaker

Future Gas SupplyFuture Gas Supply


Lower 48 unconventional gas sources will meet nearly 50% of US demand (Caruso, EIA, 2008)



Future Gas SupplyFuture Gas Supply


While tight gas sandstones represent over half of unconventional supply (Caruso, EIA, 2008)

Production Projected to Increase from Rocky Production Projected to Increase from Rocky Mountain RegionMountain Region

Tcf)

Gas

Pro

duct

ion

(


(US EIA, 2004)Date

Annu

al



Lower 48 Technically Recoverable ResourcesLower 48 Technically Recoverable Resources

peN

atur

al G

as T

y


Tcf (US EIA, 2004)

PGC Rocky Mountain Gas ResourcesPGC Rocky Mountain Gas Resources

KmvShallow Resources (0Shallow Resources (0--15,000 ft)15,000 ft) 99,167 Bcf99,167 BcfDeep Resources (15,000Deep Resources (15,000--30,000 ft)30,000 ft) 24,429 Bcf24,429 Bcf

Total Traditional ResourcesTotal Traditional Resources 123,596 Bcf123,596 BcfCoalbed Gas ResourcesCoalbed Gas Resources 63,273 Bcf63,273 BcfTotal Recoverable ResourcesTotal Recoverable Resources 186,869 Bcf186,869 Bcf


Data source: Potential Gas Committee (2003)



Why pick the Mesaverde?Why pick the Mesaverde?Tight gas sandstones (TGS) represent Tight gas sandstones (TGS) represent

72% (342 TCF) of the projected unconventional gas 72% (342 TCF) of the projected unconventional gas resource (474 TCF). resource (474 TCF). Rocky Mountain TGS are 70% of the total TGS resource Rocky Mountain TGS are 70% of the total TGS resource base (241 Tcf; USEIA 2004)base (241 Tcf; USEIA 2004)base (241 Tcf; USEIA, 2004) base (241 Tcf; USEIA, 2004) and the Mesaverde Group represents the main gas and the Mesaverde Group represents the main gas productive sandstone unit in the Rocky Mtn. TGS basinsproductive sandstone unit in the Rocky Mtn. TGS basinsand the largest shallow (<15,000 ft) target. and the largest shallow (<15,000 ft) target.

Understanding of reservoir properties and accurate Understanding of reservoir properties and accurate tools for formation evaluation are needed for:tools for formation evaluation are needed for:

assessment of the regional gas resourceassessment of the regional gas resourcej ti f f t lj ti f f t l


projection of future gas supplyprojection of future gas supplyexploration programsexploration programsoptimizing development programsoptimizing development programs

Project objectivesProject objectivesThe project provides petrophysical tools that The project provides petrophysical tools that address fundamental questions concerning address fundamental questions concerning

gas flow critical gas saturation Sgc=gas flow critical gas saturation Sgc=ff (lithofacies(lithofaciesgas flow, critical gas saturation, Sgcgas flow, critical gas saturation, Sgc ff (lithofacies, (lithofacies, Pc, architecture)Pc, architecture)capillary pressure, Pc=capillary pressure, Pc=ff (P), Pc=(P), Pc=f f (lithofacies, k, (lithofacies, k, φφ, , architecture)architecture)electrical properties, m* & n*electrical properties, m* & n*facies and upscaling issuesfacies and upscaling issues


wireline log interpretation algorithmswireline log interpretation algorithmsproviding a webproviding a web--accessible database of advanced accessible database of advanced rock properties. rock properties.



Specific research objectivesSpecific research objectivesexplore nature of critical gas saturation, capillary explore nature of critical gas saturation, capillary pressure, and electrical properties of Mesaverde pressure, and electrical properties of Mesaverde tight gas sandstonestight gas sandstonesh d th ith it bilit dh d th ith it bilit dhow do these vary with porosity, permeability, and how do these vary with porosity, permeability, and lithofacies?lithofacies?better understanding of minimum gas saturation better understanding of minimum gas saturation required for gas flowrequired for gas flowimprove log calculations through better corrections improve log calculations through better corrections for conductive solids/surface effectsfor conductive solids/surface effectsaddress the lack of adequate public domainaddress the lack of adequate public domain


address the lack of adequate public domain address the lack of adequate public domain databases covering petrophysics of tight gas databases covering petrophysics of tight gas sandstonessandstones

lots of proprietary data out there, numerous publications lots of proprietary data out there, numerous publications with partial datasets, but nothing integrated to work withwith partial datasets, but nothing integrated to work with

TasksTasksTask 1. Research Management PlanTask 1. Research Management PlanTask 2. Technology Status AssessmentTask 2. Technology Status AssessmentTask 3. Acquire Data and MaterialsTask 3. Acquire Data and Materials

Subtask 3.1. Compile published advanced properties dataSubtask 3.1. Compile published advanced properties dataSubtask 3.2. Compile representative lithofacies core and logs from major basinsSubtask 3.2. Compile representative lithofacies core and logs from major basinsSubtask 3.3. Acquire logs from sample wells and digitizeSubtask 3.3. Acquire logs from sample wells and digitize

Task 4. Measure Rock PropertiesTask 4. Measure Rock PropertiesppSubtask 4.1. Measure basic properties (k, Subtask 4.1. Measure basic properties (k, φφ, GD) and select advanced population, GD) and select advanced populationSubtask 4.4. Measure critical gas saturationSubtask 4.4. Measure critical gas saturationSubtask 4.3. Measure inSubtask 4.3. Measure in--situ and routine capillary pressuresitu and routine capillary pressureSubtask 4.4. Measure electrical propertiesSubtask 4.4. Measure electrical propertiesSubtask 4.5. Measure geologic and petrologic propertiesSubtask 4.5. Measure geologic and petrologic propertiesSubtask 4.6. Perform standard logs analysisSubtask 4.6. Perform standard logs analysis

Task 5. Build Database and WebTask 5. Build Database and Web--based Rock Catalogbased Rock CatalogSubtask 5.1. Compile published and measured data into Oracle databaseSubtask 5.1. Compile published and measured data into Oracle databaseSubtask 5.2. Modify existing webSubtask 5.2. Modify existing web--based software to provide GUI data accessbased software to provide GUI data access

Task 6. Analyze WirelineTask 6. Analyze Wireline--log Signature and Analysis Algorithmslog Signature and Analysis AlgorithmsSubtask 6 1 Compare log and core propertiesSubtask 6 1 Compare log and core properties


Subtask 6.1. Compare log and core propertiesSubtask 6.1. Compare log and core propertiesSubtask 6.2. Evaluate results and determine logSubtask 6.2. Evaluate results and determine log--analysis algorithm inputsanalysis algorithm inputs

Task 7. Simulate ScaleTask 7. Simulate Scale--dependence of Relative Permeabilitydependence of Relative PermeabilitySubtask 7.1. Construct basic bedform architecture modelsSubtask 7.1. Construct basic bedform architecture modelsSubtask 7.2. Perform numerical simulation of flow for basic bedform architectureSubtask 7.2. Perform numerical simulation of flow for basic bedform architecture

Task 8. Technology TransferTask 8. Technology Transfer



Research strategyResearch strategycompile all available published advanced compile all available published advanced rock properties (Pc, FRF, Krg, rock properties (Pc, FRF, Krg, compressibility, etc.)compressibility, etc.)compressibility, etc.)compressibility, etc.)collect 300+ core plug samples from 20 to collect 300+ core plug samples from 20 to 25 wells across 5 major basins25 wells across 5 major basinssample full range of rock types, porosity and sample full range of rock types, porosity and permeability found in Mesaverde throughout permeability found in Mesaverde throughout the Rockiesthe Rockies


the Rockiesthe RockiesKmv is widespread, lots of core available, Kmv is widespread, lots of core available, representative example for most TGS problemsrepresentative example for most TGS problems

SamplingSampling

44 wells in 6 44 wells in 6 basinsbasinsdescribeddescribed

Wind River

PowderRiver

Wyomingdescribed described 7000 ft core 7000 ft core (digital)(digital)2200 core 2200 core samplessamples120120--400 400 advanced advanced

titi

Green River

Washakie

Utah

N


properties properties samplessamples

PiceanceUintaColorado

Utah



Number of wells by basinNumber of wells by basin

8

10

12

Wel

ls Industry-contributionUSGS Core Library

0

2

4

6

8

n r

Num

ber

of W


Gre

enRi

ver

Pice

ance

Pow

der

Rive

r

Uin

ta

Was

haki

e

Was

hakie

(San

dW

ash)

Win

d Ri

ver

Basin

Core Plugs by BasinCore Plugs by Basin

500

600

700

e Pl

ugs

0

100

200

300

400

r e a e r r

Num

ber o

f Cor

e


Gre

ater

Gre

en R

ive r

Was

haki

e

Uin

ta

Pic

eanc

e

Win

d R

iver

Pow

der

Riv

er

Basin



Sampling by depthSampling by depth

Depth Histogram

0 160.180.20

80%90%100%

0 020.040.060.080.100.120.140.16

Frac

tion

10%20%30%40%50%60%70%80%


0.000.02

1000

2000

3000

4000

5000

6000

7000

8000

9000

1000

011

000

1200

013

000

1400

015

000

1600

017

000

Depth (ft)

0%10%

Property Property distributionsdistributions

Petrophysical property Petrophysical property distributions are generally distributions are generally normal or lognormal or log--normalnormalS bS b di t ib tidi t ib ti ff 5

10

15

20

25

30

35

40

45

50

Perc

ent o

f Pop

ulat

ion

(%)

AllGreen RiverPiceancePowder RiverSand WashUintahWind RiverWashakie

SubSub--distributions = distributions = f f (basin, lithofacies, (basin, lithofacies, marine/nonmarine/non--marine, etc.)marine, etc.)

30

40

50

60

sin

Popu

latio

n

Green RiverPiceancePowder RiverUintahWind River

0

5

1E-7

- 1E

-6

1E-6

- 1E

-5

1E-5

- 1E

-4

0.00

01-0

.001

0.00

1-0.

01

0.01

-0.1

0.1-

1

1-10

10-1

00

100-

1,00

0

In situ Klinkenberg Permeability (mD)

P

25

30

35

40

45

ulat

ion

(%)

AllGreen RiverPiceancePowder RiverSand WashUintah


0

10

20

30

2.58-2.60

2.60-2.62

2.62-2.64

2.64-2.66

2.66-2.68

2.68-2.70

2.70-2.72

2.72-2.74

Grain Density (g/cc)

Perc

ent o

f Bas Washakie

Sand Wash

0

5

10

15

20

0-2

2-4

4-6

6-8

8-10

10-1

2

12-1

4

14-1

6

16-1

8

18-2

0

20-2

2

22-2

4

In situ Porosity (%)

Perc

ent o

f Pop

u UintahWind RiverWashakie



Core descriptionCore description

rock typing at 0.5 ft rock typing at 0.5 ft frequency to match frequency to match q yq ylog data resolutionlog data resolutionlithology, color, grain lithology, color, grain size, sed structuressize, sed structuressample locationssample locationsimportant cementsimportant cementsd iti ld iti l


depositional depositional environmentsenvironments

Digital core descriptionDigital core description

To provide lithologic input to To provide lithologic input to equations and predict equations and predict lithology from logs used 5lithology from logs used 5lithology from logs used 5 lithology from logs used 5 digit systemdigit system

1 basic type (Ss, Ls, coal)1 basic type (Ss, Ls, coal)2 grain size/sorting/texture2 grain size/sorting/texture3 consolidation3 consolidation4 sedimentary structure4 sedimentary structure5 cement mineralogy5 cement mineralogy

P t tiP t ti tt


Property continuum Property continuum -- not not mnemonic or substitution mnemonic or substitution ciphercipherSimilar to system used in Similar to system used in our 1994 and subsequent our 1994 and subsequent studiesstudies



PetrographyPetrography

~150 advanced ~150 advanced properties smpls wereproperties smpls were

40X

properties smpls were properties smpls were petrographically petrographically characterizedcharacterizedrepresentative photos at representative photos at several magnificationsseveral magnificationspoint countspoint counts


Williams PA 424, 6148.8’ 152769.9% 2.66 g/cc Ka=0.0237 mD

100X

Core analysis programCore analysis programGeologic description of cores and rock types Geologic description of cores and rock types (Webb)(Webb)WireWire--line log analysis of all project wells over Kmv line log analysis of all project wells over Kmv (Krygowski and Whittaker)(Krygowski and Whittaker)Collect plugs for basic properties (minimum 300 Collect plugs for basic properties (minimum 300 samples, we actually collected ~2200) (Byrnes)samples, we actually collected ~2200) (Byrnes)

routine porosity and permeabilityroutine porosity and permeabilityporosity and permeability at reservoir stressporosity and permeability at reservoir stressgrain densitygrain density


Select a subSelect a sub--set of 120set of 120--400 samples for advanced 400 samples for advanced core analyses (Byrnes)core analyses (Byrnes)



“Routine” core analysis“Routine” core analysis

Routine porosity and permeabilityRoutine porosity and permeabilityInIn--situ porosity and permeabilitysitu porosity and permeabilityP l ibilit (113 l )P l ibilit (113 l )Pore volume compressibility (113 smpls)Pore volume compressibility (113 smpls)

200200--4000 psi NCS4000 psi NCSdetermined new equations fordetermined new equations for

Klinkenberg correctionKlinkenberg correctionstress dependent porositystress dependent porosityt d d t bilitt d d t bilit


stress dependent permeabilitystress dependent permeability

1

10

100

rmea

bilit

y Council GroveMesaverde/Frontier

Prior workPrior work

0.0001

0.001

0.01

0.1

n si

tu K

linke

nber

g Pe

r(m

d)


0.000010.001 0.01 0.1 1 10 100

Routine Air Permeability (md)

In

logkik = 0.0588 (logkair)3 –0.187 (logkair)2 +1.154 logkair - 0.159



SCAL workSCAL workroutine and routine and in situin situ mercury capillary pressure mercury capillary pressure investigate Pc as function of lithology,investigate Pc as function of lithology, φφ, K , K

sample span range of basins, K, lithologysample span range of basins, K, lithology

investigate stress sensitivity of Pcinvestigate stress sensitivity of Pcmost MICP curves are run under lab conditionsmost MICP curves are run under lab conditionswe expect Pc to be confining stress sensitivewe expect Pc to be confining stress sensitive120 “high120 “high--low” pairs of plugs run using highly similar plugs low” pairs of plugs run using highly similar plugs selected from selected from φφ--K dataK data

look at relationship between initial saturation and look at relationship between initial saturation and


residual gas saturation (“scanning curves”)residual gas saturation (“scanning curves”)only published data are for conventional rocksonly published data are for conventional rocksran mercury curves for this projectran mercury curves for this project

Mesaverde, Frontier capillary Mesaverde, Frontier capillary pressure vs. permeabilitypressure vs. permeability

300

350

r (ft)

10 md1 md0.1 md0.01 md

100

150

200

250

t abo

ve F

ree

Wat

er

0.01 md0.001 md


0

50

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0Water Saturation (fraction)

~Hei

ght



Pc hysteresisPc hysteresis

NonNon--wetting residual wetting residual saturation to saturation to

Drainage-ImbibitionCycles

23

4

5

imbibition Snwr = imbibition Snwr = f f (Snwi)(Snwi)this was a “freebie” this was a “freebie” added to the project added to the project planplan

1

Midale Dol

AAPG ACE 2009: Denver Colorado 29(after Larson & Morrow, 1981)

φ = 23%

SCAL workSCAL workroutine and routine and in situin situ mercury capillary mercury capillary pressurepressuredrainage critical gas saturationdrainage critical gas saturationdrainage critical gas saturationdrainage critical gas saturation




Why is Sgc important?Why is Sgc important?

0.1

1

bilit

y

P = 1.7Sgc = f (kik)

0.0001

0.001

0.01

Gas

Rel

ativ

e Pe

rmea

P = f (kik)Sgc = 10%


2 alternative views of what happens at high 2 alternative views of what happens at high Sw, which is correct?Sw, which is correct?

0.000010 10 20 30 40 50 60 70 80 90 100

Water Saturation

Saturation at capillary equilibrium for Saturation at capillary equilibrium for breakthrough pressure (Hg experiment)breakthrough pressure (Hg experiment)

60

Pc

20

30

40

50

atio

n at

Bre

akth

roug

h in

Eq

uilib

rium

(%)


0

10

0 10 20 30 40 50 60Critical Saturation at Breakthrough (%)

Satu

r

proof of concept dataset, 2005



SCAL workSCAL workroutine and routine and in situin situ mercury capillary mercury capillary pressurepressuredrainage critical gas saturationdrainage critical gas saturationdrainage critical gas saturationdrainage critical gas saturationcementation and saturation exponentscementation and saturation exponentscation exchange capacity using multication exchange capacity using multi--salinity salinity methodmethod


When F and When F and φφ are plotted logare plotted log--loglog

1000 m= 3m= 2

but not this!

F10

100m= 1

We’ve seen this before,

but not this!


φlog F = -m log φ

10.01 0.1 1



ProductsProductswebweb--based database with output as XLS files, based database with output as XLS files, graphical output, reports and presentationsgraphical output, reports and presentations

organized by data type and by area, wellorganized by data type and by area, wellhtt // k k d / d /htt // k k d / d /http://www.kgs.ku.edu/mesaverde/http://www.kgs.ku.edu/mesaverde/http://www.discoveryhttp://www.discovery--group.com/projects_doe.htmgroup.com/projects_doe.htm

methods for improved log calculationsmethods for improved log calculationsindustry talks, short courses, & forthcoming industry talks, short courses, & forthcoming publicationspublicationsso here we go..........so here we go..........



Webb: Lithofacies and Reservoir Quality

Influence of Lithofacies and DiagenesisInfluence of Lithofacies and Diagenesison Reservoir Quality of the Mesaverdeon Reservoir Quality of the Mesaverdeon Reservoir Quality of the Mesaverde on Reservoir Quality of the Mesaverde

Group, Piceance Basin, ColoradoGroup, Piceance Basin, Colorado

John WebbJohn WebbDisco er Gro p Den er CODisco er Gro p Den er CO

Denver, Colorado

Discovery Group, Denver, CODiscovery Group, Denver, CO

AAPG Short Course no. 1, Denver, COAAPG Short Course no. 1, Denver, CO

June 6, 2009June 6, 2009

11

OutlineOutline

Data collection procedures and methodsData collection procedures and methodsDi it l k l ifi ti tDi it l k l ifi ti tDigital rock classification systemDigital rock classification systemThin section preparation and petrographyThin section preparation and petrographyExample from the Piceance basinExample from the Piceance basinPaleogeography and depositional environmentsPaleogeography and depositional environmentsLithofacies and porosity/permeability relationshipsLithofacies and porosity/permeability relationshipsDetrital composition and diagenesisDetrital composition and diagenesisPorosity distributionPorosity distributionInfluence of diagenesis on reservoir qualityInfluence of diagenesis on reservoir quality

2



AcknowledgementsAcknowledgementsIndustry Partners:Industry Partners:

Bill Barrett Corporation Bill Barrett Corporation -- Steve CumellaSteve Cumella

EnCana USA, Piceance Teams EnCana USA, Piceance Teams -- Brendan Curran, Brendan Curran, Mike Dempsey, Danielle Mike Dempsey, Danielle StricklerStrickler

ExxonMobil, Piceance Basin TeamExxonMobil, Piceance Basin TeamDonDon YurewiczYurewicz HollieHollie KelleherKelleherDon Don YurewiczYurewicz, , HollieHollie Kelleher Kelleher

Williams Production Williams Production -- Lesley EvansLesley Evans

3

AcknowledgementsAcknowledgementsContractors and Government:Contractors and Government:

ElitigraphicsElitigraphics –– Peter Peter HutsonHutson

Triple O Slabbing Triple O Slabbing -- Butch OliverButch Oliver

USGS Personnel USGS Personnel -- Phil Nelson, Mark KirschbaumPhil Nelson, Mark Kirschbaum

USGS C R h C tUSGS C R h C tUSGS Core Research CenterUSGS Core Research CenterTom Michalski, Betty Adrian (current director)Tom Michalski, Betty Adrian (current director)Jeannine Honey, John Rhodes, Josh Hicks, Jeannine Honey, John Rhodes, Josh Hicks, Terri Huber, Richard Nunn, Devon Terri Huber, Richard Nunn, Devon ConnelyConnely

4



Core sampling and descriptionCore sampling and description

Cut 1” diameter plugs from butt portions of Cut 1” diameter plugs from butt portions of slabbedslabbedcore, using water cooled diamond drill bitcore, using water cooled diamond drill bitLocation of core plugs to 0.1 footLocation of core plugs to 0.1 footDigital rock typing of each core plug (lithology, grain Digital rock typing of each core plug (lithology, grain size, porosity, sedimentary structures, cementation)size, porosity, sedimentary structures, cementation)Scanned core slab images and handScanned core slab images and hand--held digital held digital photos for core plug locations and documentation of photos for core plug locations and documentation of lithology and sedimentary structureslithology and sedimentary structurest o ogy a d sed e ta y st uctu est o ogy a d sed e ta y st uctu esCore descriptions from slabbed core when possibleCore descriptions from slabbed core when possible

5

Core sampling and descriptionCore sampling and description

Logged lithology, grain size, matrix porosity, Logged lithology, grain size, matrix porosity, sedimentary structures, fractures, trace fossils, sedimentary structures, fractures, trace fossils, contact relationships and digital rock type atcontact relationships and digital rock type atcontact relationships and digital rock type at contact relationships and digital rock type at minimum ½ foot intervalsminimum ½ foot intervalsComparator for grain size determinationComparator for grain size determinationHClHCl for identification of calcareous cementsfor identification of calcareous cementsLegacy core analysis data and whole core Legacy core analysis data and whole core photographs on file at USGS CRC or from current photographs on file at USGS CRC or from current well operatorswell operatorswell operatorswell operators

6



Barrett Last Dance 43C Barrett Last Dance 43C –– Typical Core ChartTypical Core Chart

7

Digital Core Digital Core DescriptionDescription

Sampling designed to Sampling designed to sample across allsample across allsample across all sample across all lithofacieslithofacies5 digit system5 digit system

basic type (Ss, Ls, coal)basic type (Ss, Ls, coal)grain size/sorting/texturegrain size/sorting/textureConsolidation/porosityConsolidation/porositysedimentary structuresedimentary structurecement mineralogycement mineralogy

Provides lithology Provides lithology log log traces and quantitative traces and quantitative variables for multivariate variables for multivariate analysisanalysis

8



Digital Rock TypesDigital Rock Types

10xxx Shale11xxx Silty shale12xxx V shaly sandstone

Grain size/sorting/ shaliness Visible porosity

xx0xx 0-2%, unfracturedxx1xx 0-2% fractured

2 3 10% f ’d12xxx V shaly sandstone,siltstone

13xxx Shaly sandstone14xxx VF sandstone15xxx F sandstone16xxx M sandstone17xxx C sandstone18xxx VC/Matrix

supported cgl.

xx2xx 3-10%, unfrac’dxx3xx 3-10%, frac’dxx4xx 3-10%, highly fracxx5xx >10%, unfrac’dxx6xx >10%, frac’dxx7xx >10%, unfrac’dxx8xx V high, weak

consolidationxx9xx Unconsolidatedpp g

19xxx Conglomerate

05000 Volcanic ash2xxxx Limestone30000 Coal

xx9xx Unconsolidated

Porosity/ Resistivity logs

GR/Porosity/ Resistivity logs9

Digital Rock Types, cont.Digital Rock Types, cont.

Cementxxxx0 Pyrite

1 Sid i

Sedimentary struc’sxxx0x Vertical dikexxx1x Bioturbatedxxx2x Contorted xxxx1 Siderite

xxxx2 Phosphatexxxx3 Anhydrite xxxx4 Dolomitexxxx5 Calcitexxxx6 Quartzxxxx7 Authigenic clayxxxx8 Carbonaceousxxxx9 No pore filling

xxx2x Contortedxxx3x Discontinuous

laminationsxxx4x Continuous

laminationsxxx5x Flaser beddedxxx6x Ripple laminatedxxx7x Trough & planar

tabular crossbeds xxxx9 No pore fillingDensity/ Resistivity/ PE logs

10

xxx8x Planar laminated, low angle cross bedded

xxx9x Massive beddedShaliness, vertical and lateral permeability



15277 - Medium sandstone withmoderate porosity, not fractured,trough cross bedded,clay cementedclay cemented

11

Utility of digital rock typing, continuedUtility of digital rock typing, continued

Excellent match with GR log traces, core gammaExcellent match with GR log traces, core gammaPrecise depth shifting of core analysis dataPrecise depth shifting of core analysis dataD t t i fl f i i d h liD t t i fl f i i d h liDemonstrates influence of grain size and shaliness on Demonstrates influence of grain size and shaliness on porosity and permeabilityporosity and permeabilityAllowed improvement of equations used to calculate Allowed improvement of equations used to calculate Archie Archie SwSw, total and effective porosity and significantly , total and effective porosity and significantly improved estimates of permeabilityimproved estimates of permeabilityRock types are not restricted to a specific depositional Rock types are not restricted to a specific depositional environmentenvironmentLog analysis identified detrital shale component, but Log analysis identified detrital shale component, but failed to identify details of grain size and sedimentary failed to identify details of grain size and sedimentary structuresstructures

12



Correlation of lithofacies and core Correlation of lithofacies and core analysis data to analysis data to wirelinewireline logslogs

13

Utility of digital rock typingUtility of digital rock typing

Track statistical distribution of lithofacies for Track statistical distribution of lithofacies for sampling and core analysis datasampling and core analysis dataProvides quantitative variables for multivariate Provides quantitative variables for multivariate analysisanalysisThe simple variation in grain density from basin to The simple variation in grain density from basin to basin indicates that differences in detrital basin indicates that differences in detrital composition of sediment, depositional environment, composition of sediment, depositional environment, burial history and diagenesis among basins burial history and diagenesis among basins requires separate treatment of basins for requires separate treatment of basins for assessment of reservoir qualityassessment of reservoir quality

14



Grain densities of the Grain densities of the Mesaverde GroupMesaverde Group

60

atio

nGreen River

10

20

30

40

50en

t of B

asin

Pop

ula Green River

PiceancePowder RiverUintahWind RiverWashakieSand Wash

0

10

2.58-2.60

2.60-2.62

2.62-2.64

2.64-2.66

2.66-2.68

2.68-2.70

2.70-2.72

2.72-2.74


Perc

15

Thin section preparationThin section preparationBlueBlue--dyed epoxy, low viscosity, slow curedyed epoxy, low viscosity, slow cureVacuum and pressure impregnation in warm Vacuum and pressure impregnation in warm ovenovenovenovenPolished surfaces of billet and mounted Polished surfaces of billet and mounted slideslideDual carbonate stained for nonferroan (red) Dual carbonate stained for nonferroan (red) and ferroan carbonate (various shades of and ferroan carbonate (various shades of blue)blue)blue)blue)Stained for potassium feldspar (KStained for potassium feldspar (K--spar is spar is yellow)yellow)Cover slipsCover slips

16



Thin section petrographyThin section petrography

Nikon and Nikon and LeitzLeitz petrographic microscopespetrographic microscopesConventional film and digital photography, Conventional film and digital photography, representative magnifications and detailed representative magnifications and detailed featuresfeatures300 point counts per sample, automated 300 point counts per sample, automated point count stagepoint count stageCalculations in Excel, graphic plots in Calculations in Excel, graphic plots in g p pg p pQuattro Pro and Excel spreadsheetsQuattro Pro and Excel spreadsheets

17

Utility of thin section petrographyUtility of thin section petrographyDetrital compositionDetrital composition

ProvenanceProvenanceRadioactive components for GR matchRadioactive components for GR matchRadioactive components for GR matchRadioactive components for GR matchBulk density of constituent grainsBulk density of constituent grains

CementsCementsBulk density of constituent cement (calcite, Bulk density of constituent cement (calcite, dolomite, pyrite, clay)dolomite, pyrite, clay)

Distribution of clayDistribution of clayDistribution of clayDistribution of clayDetrital Detrital -- laminated, structural, dispersed laminated, structural, dispersed (burrowing)(burrowing)Clay cements Clay cements –– porepore--lining, porelining, pore--bridging or bridging or disperseddispersedClay mineralogy (visual morphology)Clay mineralogy (visual morphology) 18



Utility of thin section petrographyUtility of thin section petrographyDiagenesisDiagenesis

Assess the effect of compaction and pressure Assess the effect of compaction and pressure solutionsolutionDocument changes in detrital grains or rock Document changes in detrital grains or rock fabricfabric

Porosity distributionPorosity distributionMesoporosity, microporosity, moldic and Mesoporosity, microporosity, moldic and intragranular porosityintragranular porosityCompare relative abundance of Meso vs. MicroCompare relative abundance of Meso vs. Micro

FracturesFracturesAssess the importance of microfracturesAssess the importance of microfracturesIdentify fracture cementsIdentify fracture cements

19

Paleogeography of Mesaverde Group, Paleogeography of Mesaverde Group, Uinta and Piceance BasinsUinta and Piceance Basins

Early Clagget time, Mancos Shale Middle Judith River time,

Iles Formation (Rollins, Cozette and Corcoran Ss)

Middle Bear Paw time,Williams Fork Formation

McGookey, et al., 1972

approx 80 mya

approx 73 mya

approx 70 mya20



Depositional environments of the Depositional environments of the MesaverdeMesaverde

Shallow marine and shoreline environments, Shallow marine and shoreline environments, including lagoonal bayincluding lagoonal bay fill and coastalfill and coastalincluding lagoonal, bayincluding lagoonal, bay--fill and coastal fill and coastal marshmarshTidal delta, tidal channel, mudflat and tidally Tidal delta, tidal channel, mudflat and tidally influenced coastal streamsinfluenced coastal streamsCoal swamps (raised mire) and coastal plainCoal swamps (raised mire) and coastal plainFl i l h l i l di tid ll i fl dFl i l h l i l di tid ll i fl dFluvial channel, including tidally influencedFluvial channel, including tidally influencedAbandoned channel and overbank/splayAbandoned channel and overbank/splayPaleosolsPaleosols, rooted horizons, air fall ash and , rooted horizons, air fall ash and lacustrine to shallow marine limestonelacustrine to shallow marine limestone

21

Example: The Piceance BasinExample: The Piceance Basin

Core analysis: Core analysis: RoutineRoutine -- 629 samples, SCAL629 samples, SCAL -- 46 samples46 samplesRoutine Routine 629 samples, SCAL 629 samples, SCAL 46 samples46 samples

Mercury invasion and imbibition curves for 8 Mercury invasion and imbibition curves for 8 samplessamplesCore description and petrography : Core description and petrography :

6 wells, 2 shallow bore holes, 1168’ core, 46 thin 6 wells, 2 shallow bore holes, 1168’ core, 46 thin section point counts section point counts

L l iL l iLog analysis:Log analysis:Modern log suites for 5 wells, various vintages and Modern log suites for 5 wells, various vintages and format for format for 11 older well and 2 shallow bore holesolder well and 2 shallow bore holes

22



Mesaverde Group cores, Piceance BasinMesaverde Group cores, Piceance Basin

W Fuels 21011-5 Moon Lake White River Dome

EM WR T63X-2G

Chevron 33-34MWX-2 BBC LD 43C-3-792

Moon Lake

USGS BC 1

White River Dome

Love Ranch

Grand Valley

Parachute

RulisonMamm Creek

FR M30-2-96W WRD

Wms PA 424-34

23

Stratigraphic distribution of samples, Stratigraphic distribution of samples, Piceance BasinPiceance Basin

33-34

USGS Coal Resources, #1 Book Cliffs outcrop core

10,500 ft5700 ft4,600 ft

3,500 ft

8,100 ft

6,500 ft 6,600 ft

250 ft

6,300 ft

24

8200 ft



Barrett Last Dance 43C Barrett Last Dance 43C –– Shallow Marine/CoastalShallow Marine/Coastal

25

Barrett Last Dance 43C Barrett Last Dance 43C –– Coastal MudstonesCoastal Mudstones

26



Barrett Last Dance 43C Barrett Last Dance 43C –– FluvialFluvial

27

Lithofacies Lithofacies -- Influence of grain size and shaliness on Influence of grain size and shaliness on porosity and permeabilityporosity and permeability

100

Phi/K Crossplot Mesaverde Group, Piceance Basin

0.01

0.1

1

10

bien

t Per

mea

bilit

y, in

mD

11XXX12XXX13XXX14XXX15XXX16XXX

28

0.0001

0.001

0 5 10 15 20

Amb

Ambient Porosity, percent

16XXX17XXX



10

100m

D


0.01

0.1

1

Ambi

ent P

erm

eabi

lity,

in m

11XXX

0.0001

0.001

0 5 10 15 20

Ambient Porosity, percent29

10

100

mD


0.01

0.1

1

Ambi

ent P

erm

eabi

lity,

in m

12XXX

0.0001

0.001

0 5 10 15 20




10

100m

D


0.01

0.1

1

Ambi

ent P

erm

eabi

lity,

in m

13XXX

0.0001

0.001

0.0 5.0 10.0 15.0 20.0


10

100

mD


0.01

0.1

1

Ambi

ent P

erm

eabi

lity,

in m

14XXX

0.0001

0.001

0 5 10 15 20




10

mD


0.01

0.1

1

Ambi

ent P

erm

eabi

lity,

in m

15XXX

0.0001

0.001

0 5 10 15 20


10

mD


0.01

0.1

1

Ambi

ent P

erm

eabi

lity,

in m

16XXX17XXX

0.0001

0.001

0 5 10 15 20




10

mD


0.01

0.1

1

Ambi

ent P

erm

eabi

lity,

in m

16XXX17XXX

0.0001

0.001

0 5 10 15 20


Phi/K Crossplot Mesaverde Group, Piceance BasinFine Grained Ss (15xxx)

Influence of burial on porosity and permeability of lithofaciesInfluence of burial on porosity and permeability of lithofacies

0.1

1

10

100

bien

t Per

mea

bility, in mD

250 ‐ 3999 ft

4000 ‐ 6999 ft

7000 ‐ 10,000 ft

0.001

0.01

0 5 10 15 20 25

Amb




Phi/K Crossplot Mesaverde Group, Piceance BasinMedium Grained Ss (16xxx)

Influence of burial on porosity and permeability of lithofaciesInfluence of burial on porosity and permeability of lithofacies

0.1

1

10

100

ability, in m

D

250 ‐ 3999 ft

4000 ‐ 6999 ft

7000 ‐ 10,000 ft

0.001

0.01

0 5 10 15 20 25

Ambien

t Per

mea


Detrital Composition of SandstonesDetrital Composition of Sandstonesin the Mesaverde Groupin the Mesaverde GroupWhy do we care? Because detrital composition has Why do we care? Because detrital composition has an effect on diagenesis and porosity preservation.an effect on diagenesis and porosity preservation.

In the Mesaverde, quartzose sandstones are In the Mesaverde, quartzose sandstones are preferentially subject to pressure solution preferentially subject to pressure solution compaction and quartz overgrowth cementation compaction and quartz overgrowth cementation (clay cementation may retard overgrowths)(clay cementation may retard overgrowths)

Feldspathic sandstones suffer compaction by grain Feldspathic sandstones suffer compaction by grain rearrangement and brittle rearrangement and brittle deformation, accompanied by clay cement. deformation, accompanied by clay cement.

38



Detrital Composition of SandstonesDetrital Composition of Sandstonesin the Mesaverde Groupin the Mesaverde Group

Other alterations include dissolution of framework Other alterations include dissolution of framework grains (Kgrains (K--spar and carbonate rock spar and carbonate rock g (g ( ppfragments), resulting in moldic porosity.fragments), resulting in moldic porosity.

Ductile deformation of shale, carbonaceous Ductile deformation of shale, carbonaceous material, volcanic rock fragments and micaceous material, volcanic rock fragments and micaceous grains, brittle deformation of feldsparsgrains, brittle deformation of feldspars

39

Detrital Composition of SandstonesDetrital Composition of Sandstonesin the Mesaverde Groupin the Mesaverde Group

Composition ranges from litharenite to feldspathic Composition ranges from litharenite to feldspathic litharenite lithic arkose sublitharenitelitharenite lithic arkose sublitharenite subarkosesubarkoselitharenite, lithic arkose, sublitharenite, litharenite, lithic arkose, sublitharenite, subarkosesubarkoseand quartzareniteand quartzareniteRock fragments include volcanic, sedimentary and Rock fragments include volcanic, sedimentary and metamorphic grainsmetamorphic grainsVolcanic rock fragments are commonly Volcanic rock fragments are commonly altered, resulting in replacement by altered, resulting in replacement by clayclay silicificationsilicification and partial to complete dissolutionand partial to complete dissolutionclay, clay, silicificationsilicification and partial to complete dissolutionand partial to complete dissolutionSedimentary rock fragments include Sedimentary rock fragments include shale/mudstone, chert and carbonate grainsshale/mudstone, chert and carbonate grains

40



41

42



43

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

3544.9

3555.4

Detrital Composition, Barrett Last Dance 43C

Williams Fork Fm

3577.6

4004.3

4013.3

4393.6

4416.6

Top Gas 4363 ft

44

5715.4

6042.4

6337.1

Quartz Feldspar Lithic

Cameo Coal zone



0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

5734.1

5838.6

5852.3

6536.3

Detrital Composition, MWX‐2

Williams Fork Fm

6542.2

6550.3

7085.5

7133.5

7264.5

7272.8

7276.2Cozette Ss

45

7851.3

7877.5

7880.1

8106.9

8117.9

Quartz Feldspar Lithic

Corcoran Ss

Cozette Ss

SRF – Sedimentary rock fragmentsVRF V l i k f tVRF – Volcanic rock fragmentsPRF – Plutonic rock fragmentsQM – Quartzose metamorphic MRF – Micaceous metamorphic

46



0.0 5.0 10.0 15.0 20.0 25.0

3544.9

3555.4

Lithic Population, Barrett Last Dance 43C

Williams Fork Fm

3577.6

4004.3

4013.3

4393.6

4416.6

Top Gas 4363 ft

47

5715.4

6042.4

6337.1

Chert Shale Dolostone Volcanic

Cameo Coal zone

0 5 10 15 20 25

5734.1

5838.6

5852.3

6536.3

Lithic Population, MWX‐2

Williams Fork Fm

6542.2

6550.3

7085.5

7133.5

7264.5

7272.8

7276.2Cozette Ss

48

7851.3

7877.5

7880.1

8106.9

8117.9

Chert Shale Limestone Dolostone Volcanic

Corcoran Ss

Cozette Ss



Cement Distribution in the Mesaverde GroupCement Distribution in the Mesaverde Group

PorePore--lining clay cementslining clay cementsChlorite (common to abundant)Chlorite (common to abundant)MixedMixed--layer illitelayer illite--smectite (sparse to moderate)smectite (sparse to moderate)

PorePore--filling cementsfilling cementsSiderite (trace)Siderite (trace)Pyrite (trace to sparse)Pyrite (trace to sparse)NonNon--ferroan calcite (sparse)ferroan calcite (sparse)Quartz overgrowth (trace to abundant)Quartz overgrowth (trace to abundant)Ferroan calcite and ferroan dolomite (sparse to common)Ferroan calcite and ferroan dolomite (sparse to common)Albite (grain replacement and moldAlbite (grain replacement and mold--filling)filling)Kaolinite (sparse in one sample in Book Cliff outcrop)Kaolinite (sparse in one sample in Book Cliff outcrop)

49

0 5 10 15 20 25

3544.9

3555.4

Cement Types, Barrett Last Dance 43C

Williams Fork Fm

3577.6

4004.3

4013.3

4393.6

4416.6

Top Gas 4363 ft

50

5715.4

6042.4

6337.1

Quartz Og Fe Calcite Chlorite and ML/IS

Cameo Coal zone



0 5 10 15 20 25 30 35

5734.1

5838.6

5852.3

6536.3

Cement Types, MWX‐2

Williams Fork Fm

6542.2

6550.3

7085.5

7133.5

7264.5

7272.8

7276.2Cozette Ss

7851.3

7877.5

7880.1

8106.9

8117.9

Quartz Og Fe Calcite Chlorite and ML/IS

Cozette Ss

Corcoran Ss

51

Porosity Distribution in the Mesaverde GroupPorosity Distribution in the Mesaverde Group

MesoporosityMesoporosityPore throat apertures <2 micron, > 0.5 micron radiusPore throat apertures <2 micron, > 0.5 micron radiusIntergranular pores, primary and secondary Intergranular pores, primary and secondary Moldic pores (partly and completely dissolved Moldic pores (partly and completely dissolved feldspars, carbonate and volcanic rock fragments (large feldspars, carbonate and volcanic rock fragments (large aspect ratio, pore body/pore throat)aspect ratio, pore body/pore throat)

MicroporosityMicroporosityPore throat apertures <0.5 micron, >0.1 micron radius Pore throat apertures <0.5 micron, >0.1 micron radius PP li i dli i d filli l tfilli l tPorePore--lining and porelining and pore--filling clay cementfilling clay cementIntragranular micropores (altered VRF, clay pellets, shale Intragranular micropores (altered VRF, clay pellets, shale rock fragments, clay and carbonaceous matrix)rock fragments, clay and carbonaceous matrix)

52



Porosity Distribution in the Mesaverde GroupPorosity Distribution in the Mesaverde Group

NanoporosityNanoporosityPore throat apertures <0.1 micron radiusPore throat apertures <0.1 micron radiusTypical of mudstones, clayTypical of mudstones, clay--sized intergranular, common in sized intergranular, common in d t it l l b t i ld t it l l b t i ldetrital clay or carbonaceous materialdetrital clay or carbonaceous material

FracturesFracturesMacroscopic Macroscopic Microscopic (primarily crushed feldspars or chert, partings Microscopic (primarily crushed feldspars or chert, partings or separations at quartz overgrowth boundaries)or separations at quartz overgrowth boundaries)

53

54Interparticle and intercrystalline Mesoporosity

Interparticle and intraparticleMicroporosity



0 5 10 15 20 25

3544.9

3555.4

Porosity Distribution, Barrett Last Dance 43C

Williams Fork Fm

3577.6

4004.3

4013.3

4393.6

4416.6

Top Gas 4363 ft

55

5715.4

6042.4

6337.1

BP sBP Mo clfBP

Cameo Coal zone

0 2 4 6 8 10 12

5734.1

5838.6

5852.3

6536.3

Porosity Distribution, MWX‐2

Williams Fork Fm

6542.2

6550.3

7085.5

7133.5

7264.5

7272.8

7276.2Cozette Ss

7851.3

7877.5

7880.1

8106.9

8117.9

BP sBP Mo clfBP

Corcoran Ss

Cozette Ss

56



57

Porosity Networks in the Mesaverde GroupPorosity Networks in the Mesaverde Group

Type IType IConventional porosity Conventional porosity –– Primary intergranular and Primary intergranular and modified intergranular (e.g. quartz overgrowth modified intergranular (e.g. quartz overgrowth cement, secondary intergranular)cement, secondary intergranular)Lacking clay cementLacking clay cementMesoporosity >> MicroporosityMesoporosity >> MicroporosityPhi=high, K=high, low Swi, efficient drainage, low to Phi=high, K=high, low Swi, efficient drainage, low to moderate Pc entry pressuremoderate Pc entry pressure

Type IIType IIIntergranular and moldic Intergranular and moldic –– May include primary May include primary intergranular and secondary intergranular intergranular and secondary intergranular Trace to absent clay cementTrace to absent clay cementMesoporosity >> MicroporosityMesoporosity >> MicroporosityPhi=high, K=moderate , low to moderate Swi, elevated Phi=high, K=moderate , low to moderate Swi, elevated SrgSrg, moderate Pc entry pressure, moderate Pc entry pressure 58




Type IIIType IIIRestricted intergranular Restricted intergranular –– ClayClay--lined pores and pore lined pores and pore throats, some moldic and claythroats, some moldic and clay--filled intergranular filled intergranular microporositymicroporositymoderate to common clay cementmoderate to common clay cementMicroporosity Microporosity >> MesoporosityMesoporosityPhi=moderate, K=low, moderate to high Swi, elevated Phi=moderate, K=low, moderate to high Swi, elevated SrgSrg, increased Pc entry pressure, increased Pc entry pressure

Type IVType IVMicrointergranularMicrointergranular –– ClayClay--filled intergranular poresfilled intergranular poresModerate to common clay cementModerate to common clay cementMicroporosity >> MesoporosityMicroporosity >> MesoporosityHigh Swi, Phi=moderate to low, K=low to extremely High Swi, Phi=moderate to low, K=low to extremely low, elevated low, elevated SrgSrg, increased Pc entry pressure, increased Pc entry pressure

59


Type VType VNanointergranularNanointergranular–– Typical of mudstones, clayTypical of mudstones, clay--sized sized intergranular, common clay or carbonaceous materialintergranular, common clay or carbonaceous materialMicroporosity onlyMicroporosity onlyPhi=moderate to low, K=low to extremely low, high Phi=moderate to low, K=low to extremely low, high Swi, extremely high pore entry pressureSwi, extremely high pore entry pressure

60



Type I (shallow burial)

Porosity consists of well connected primary and secondary intergranular mesopores, sparse moldic pores, quartz overgrowth cement.

Quartz cement is sparse.

Lack of pore-lining clay cement reduces Swi and improves relative permeability.

40X

100X

USGS CB #1 Book Cliffs, 255.8’ Rock type 15567Porosity 24.8% amb., Rhob2.64 g/cc Ka=137.62 mD Kins=112.2 mD

61

Type I (moderate burial)

Porosity consists of moderately connected primary and secondary intergranular mesopores and traces of pore-lining chlorite clay containing microporosity

40X

microporosity.

Quartz cement and ferroan calcite are sparse.

Lack of pore-lining clay cement reduces Swi and improves relative permeability.

100X

Barrett Last Dance 43C, 3544.9’ Rock type 16277Porosity 11.4% Rhob 2.65 g/cc Ka=0.8716 mD Kins=0.4287 mD

62



Type II

Porosity consists of poorly to moderately connected moldic and secondary intergranular mesopores with traces of pore-lining ML/IS(?) clay, containing microporosity.clay, containing microporosity.

Quartz cement is prominent, ferroan calcite is sparse.

Pore-lining clay cement begins to increase Swi and reduce relative permeability.

40X

Williams PA 424, 6148.8’ Rock type 15276Porosity 9.9% Rhob 2.66 g/cc Ka=0.0237 mD Kins=0.0076 mD

100X 63

Type III

Porosity consists of clay-lined intergranular pores, pore throats are occluded by clay cement, causing elevated Swi, reduced relative permeability and i d P t

40X

increased Pc entry pressure.

Cements include chlorite or ML-IS clay, traces of nonferroan or ferroan calcite, traces of quartz overgrowths.

Inhomogeneous packing and over-sized intergranular pores

100X


over sized intergranular pores indicate the development of secondary intergranular porosity.

64



Type III

Porosity consists of clay-lined intergranular pores, pore throats are occluded by clay cement, which causes elevated Swi, reduced relative permeability

d i d P t

400X

and increased Pc entry pressure

Cements include chlorite or ML-IS clay, traces of nonferroan or ferroan calcite, traces of quartz overgrowths.

Inhomogeneous packing and over-sized intergranular pores

400X, XP

over sized intergranular pores indicate the development of secondary intergranular porosity.

.

65


Type IV

Porosity consists almost entirely of sparse, poorly connected, clay-filled intergranular microporosity.

Quartz cement is prominent

40X

Quartz cement is prominent, ferroan calcite is sparse.

Pore-filling clay cement causes elevated Swi, reduced relative permeability and increased Pc entry pressure.

100X


66



Type V

Porosity consists entirely of sparse, poorly connected microporosity within interparticle voids of mudstone and shale matrix.

Cements include siderite, ferroan calcite and pyrite. Organic matter is locally common.

Abundant clay causes highly elevated Swi, severely reduced permeability and elevated Pc entry pressure.

64X

CER MWX-2, 7085.5’ Rock type 11299Porosity 2.4% Rhob 2.70 g/cc Ka=0.0020 mD Kins=0.00004 mD

p

67160X

100

n mD

Porosity types, Mesaverde, Piceance basin

0.1

1

10

Perm

eability, ambien

t, in

Type I

Type II

Type III

Type IV

Type V

0.001

0.01

0 5 10 15 20 25

P

Porosity, ambient, in percent

68



100

mD

Porosity types, Mesaverde, Piceance basin250 ‐ 3999 ft minimum burial

0.1

1

10

ermeability, ambien

t, in

m

Type I

Type II

Type III

Type IV

Type V

0.001

0.01

0 5 10 15 20 25

Pe


69

100

mD

Porosity types, Mesaverde Group, Piceance basin4,000 ‐ 6,999 ft minimum burial

0.1

1

10

ermeability, ambien

t, in

m

Type I

Type II

Type III

Type IV

Type V

0.001

0.01

0 5 10 15 20 25

Pe


70



100

mD

Porosity types, Mesaverde Group, Piceance basin7,000 ‐ 10,000 ft minimum burial

0.1

1

10

ermeability, ambien

t, in

m

Type I

Type II

Type III

Type IV

Type V

0.001

0.01

0 5 10 15 20 25

Pe


71

Diagenetic alterations in the MesaverdeDiagenetic alterations in the Mesaverde

Compaction, ductile and brittle deformationCompaction, ductile and brittle deformationClay cements, primarily chlorite and MLClay cements, primarily chlorite and ML--ISISQuartz overgrowthsQuartz overgrowthsNonferroan calciteNonferroan calciteDissolution of calcite or other precursor cementsDissolution of calcite or other precursor cementsFerroan calcite and ferroan dolomite cementsFerroan calcite and ferroan dolomite cementsReplacement of KReplacement of K--spar by ferroan calcite and spar by ferroan calcite and albite formation of moldic porosityalbite formation of moldic porosityalbite, formation of moldic porosityalbite, formation of moldic porosityDissolution of carbonate rock fragmentsDissolution of carbonate rock fragments

72



Brittle deformation of K-spar and Pore-lining clay cement – Chlorite, ferroan calcite pore fill

73

Pore-filling chlorite cement with continued burial 74



Pore-lining clay cement – ML/IS 75





Inhomogeneous packing and relics of calcite cement indicate secondary intergranular porosity

78



Relic of calcite cement and adjacent secondary intergranular porosity 79

Secondary intergranular pores mimic size and shape of neighboring cement-filled areas 80



Secondary porosity, created by dissolution of framework grains81

Secondary porosity, created by dissolution of framework grains82



Secondary porosity, created by dissolution of carbonate framework grains

83

Alteration of potassium feldspar 84



Alteration of potassium feldspar and VRF’s 85

Alteration of potassium feldspar 86



Alteration of plagioclase feldspar 87

Alteration of plagioclase feldspar 88



Alteration of volcanic rock fragments89

Influence of depositional environment on detrital composition 90



Influence of depositional environment on detrital composition 91

Influence of depositional environment on diagenesis 92



Pore-filling chlorite in a quartzose sandstone 93

Pore-filling chlorite in a quartzose sandstone 94



ConclusionsConclusionsRock typing is useful tool for lithofacies Rock typing is useful tool for lithofacies analysis and directing statistical sampling.analysis and directing statistical sampling.Grain size and shale content are the primaryGrain size and shale content are the primaryGrain size and shale content are the primary Grain size and shale content are the primary influences on reservoir qualityinfluences on reservoir qualityCompaction and cementation by clay Compaction and cementation by clay (primarily chlorite and ML(primarily chlorite and ML--IS), quartz and IS), quartz and ferroan calcite further reduce porosity and ferroan calcite further reduce porosity and permeabilitypermeabilityMatrix porosity in the Mesaverde Group Matrix porosity in the Mesaverde Group consists of both primary and secondary consists of both primary and secondary intergranular, moldic and clayintergranular, moldic and clay--filled filled microporositymicroporosity

95

Conclusions, continuedConclusions, continued

Mesofractures, microfractures on the scale of Mesofractures, microfractures on the scale of individual grains and overgrowth partings areindividual grains and overgrowth partings areindividual grains, and overgrowth partings are individual grains, and overgrowth partings are also presentalso presentPorosity type and distribution of clay cements Porosity type and distribution of clay cements help explain the variation of permeability for a help explain the variation of permeability for a given value of porositygiven value of porosityLog analysis is complicated by the presence Log analysis is complicated by the presence g y p y pg y p y pof chlorite clay cement (more on that later…)of chlorite clay cement (more on that later…)

96



Analysis of Critical Permeability, Capillary Pressure and Electrical Properties for Mesaverde Tight Gas Sandstones from Western

http://www.kgs.ku.edu/mesaverde

Gas Sandstones from Western U.S. Basins

DOE Contract DE-FC26-05NT42660

http://www.discovery-group.com

97


Byrnes: Porosity, Permeability, and Compressibility

Analysis of Critical Permeability, Capillary and Electrical Properties

for Mesaverde Tight Gas Sandstones f W t U S B i

Analysis of Critical Permeability, Capillary and Electrical Properties

for Mesaverde Tight Gas Sandstones f W t U S B ifrom Western U.S. Basins from Western U.S. Basins

US DOE # DE-FC26-05NT42660US DOE # DE-FC26-05NT42660http://www.kgs.ku.edu/mesaverdehttp://www.kgs.ku.edu/mesaverde

Core Analysis• Porosity & Grain Density

– Lithologic and other controls– Routine helium– In situ– Pore Volume Compressibility

• Saturation & Capillary Pressure– Routine Analysis (retort, Dean-Stark)– Air-brine, oil-brine, air-mercury– Drainage, imbibition– Centrifuge, Porous-plate, Hg intrusion

I t f i l T ip y

• Permeability– Routine Air– Klinkenberg– Crack & Capillary– Liquid– In situ– Effective & Relative

• Gas oil Oil water Gas water

– Interfacial Tension– Contact Angle– Wettability– Threshold Pressure

• Enhanced Oil Recovery– Chemical (polymer, surfactant, caustic)– Miscible (CO2, N2, Enriched Gas)– Thermal (Steam, Combustion)

• Electrical & Acoustic Properties• Gas-oil, Oil-water, Gas-water• Drainage, imbibition• Steady-state, unsteady-state• Single-phase stationary• Parameters influencing kr

– T, Poverburden, wettability, pore architecture, capillary number

– Fluid Sensitivity

• Electrical & Acoustic Properties – Archie Electrical Properties

• Cementation & Saturation Exponent, Cation Exchange

– Vp & Vs• Rock Mechanics

– Young’s Modulus, Poisson’s Ratio, Bulk Modulus

– Fracture Pressure



• All petrophysical properties are

• All petrophysical properties are

• X - Composition (broad definition)– Classification (sandstone,

limestone, etc.)– Compositional (mineralogy)– Textural (sorting-grain size

• X - Composition (broad definition)– Classification (sandstone,

limestone, etc.)– Compositional (mineralogy)– Textural (sorting-grain size

PVTXtproperties are physical-chemical in nature and dependent on:

• P – Pressure– Confining/pore

properties are physical-chemical in nature and dependent on:

• P – Pressure– Confining/pore

– Textural (sorting-grain size distribution, roundness, angularity)

– Sedimentologic (bedding, heterogeneity, architecture)

– Porosity/ pore size distribution

– Fluid

– Textural (sorting-grain size distribution, roundness, angularity)

– Sedimentologic (bedding, heterogeneity, architecture)

– Porosity/ pore size distribution

– Fluid• V- Volume/Scale• T – Temperature• t – time/history

(hysteresis)

• V- Volume/Scale• T – Temperature• t – time/history

(hysteresis)

FluidFluid

Always consider at what conditions a property was measured and over what range of conditions the measured property value is valid

PorosityPorosity



Core Analysis• Porosity

– Classification– Lithologic & other controls– Routine helium– In situ

l ibili


f i l i– Pore volume compressibility– Wireline-log Analysis

• Permeability– Routine Air– Klinkenberg– Crack & Capillary– Liquid– In situ– Effective & Relative


• Enhanced Oil Recovery– Chemical (polymer, surfactant,

caustic)– Miscible (CO2, N2, Enriched Gas)– Thermal (Steam, Combustion)Effective & Relative

• Gas-oil, Oil-water, Gas-water• Drainage, imbibition• Steady-state, unsteady-state• Single-phase stationary• Parameters influencing kr



( , )• Electrical & Acoustic Properties

– Archie Electrical Properties• Cementation & Saturation

Exponent, Cation Exchange– Vp & Vs

• Rock Mechanics– Young’s Modulus, Poisson’s Ratio, Bulk

Modulus– Fracture Pressure



Porosity Types - Classifications

Intraparticle φVuggy φ

Transparticle φF t φ

Various Porosity Nomenclature – genesis, size distribution, flow contribution

Nano <0.1 μmMicro 01.-0.5 μm

Micro φIneffective φ

Vuggy φSecondary φ

Fracture φ Meso 0.5-2 μmMacro 2-10 μmMega 10-100 μm

Interparticle φPrimary φEffective φ

Porosity DefinitionPorosity, n. The ratio of void space to the bulk volume of rock containing that void spacePorosity, n. The ratio of void space to the bulk volume of rock containing that void space

φi=isolatedφc=connected = φcmicro+φcmacro+φboundφ d 0

φ = Vp/(Vp+Vg)

Connected φ

Isolated φ (minor)

φcmacro= connected, >0.5μmφcmicro= connected, <0.5μm, not boundφbound = connected, bound to clay or

surface, water of hydration

• Total φtotal = φc+φi= φcmacro+φcmicro+φbound+φi

• Effective1 φ = φ (excludes φ )

micro φbound-water φ

• Effective1 φeff = φc (excludes φi)• Effective2 φeff = φcmacroi+φmicro (exc φi,

φbound)• Effective3 φeff = φcmacro+φi+φcmicro (exc

φbound)• Effective4 φeff = φcmacro (exc

φi,φcmicro,φbound)



Packing & Sorting Control on Porosity

(after Bear , 19

Porosity independent of sizeHighly dependent on sorting & packing

Secondary Porosity - Transfer

• Feldspar grain dissolution

t dcreates secondary porosity but removed material often reprecipitates in nearby pore

k li ispace as kaolinite or smectite



Porosity Measurement• Core Analysis

– Helium Boyle’s law - Dry sample, measure bulk volume, injected gas measures grain volume - measures φc, does not measure φi and may not measure some φbound

– Crushed sample He pycnometer – dry crushed sample material is measured by Boyle’s Law technique measures φBoyle s Law technique, measures φt

– Liquid Resaturation – dry sample is weighed,saturated with liquid of know density and weighed saturated, weight difference measures φc, does not measure φi and may not measure some φbound

– Summation of Fluids – two pieces of native core, one is weighed, crushed, retorted for oil&water content, and weighed; second has bulk volume measured and mercury injected into gas pore space, fluid saturations and porosity calculated for combined volumes – measures combination of φt and φc

– Nuclear Magnetic Resonance – integrated NMR signal is measured on t t d l φsaturated sample – measures φt

• Wireline Logs– Density (ρma- ρb)/ (ρma- ρliq)– Sonic (Δt- Δtma)/(Δtfluid- Δtma)– ResistivityF = a/φm

– NMR– Neutron

Core Analysis Data

Core Analysis Data

Helium Porosimeter Precision• Vg = (Vr +Vc) -P1g/P2gVr

(after Ruth & Pohjoisrinne

Properly performed error in grain volume measurement should be < +0.001 cc



Porosity Error -Interlaboratory Calibration

• Dotson et al (1951) Avg φ Error = + 0.5• Thomas and Pugh (1988) Maximum “acceptable” deviation =

+ 0 5; 65% of labs in 1987 met that quality assurance criteria+ 0.5; 65% of labs in 1987 met that quality assurance criteria• Quality reviewed data in TGS +0.25 pu (Hunt & Luffel, 1988)

Permeability Permeabilityto air, md Ambient Overburden to air, md Ambient Overburden

Xmean 248 19.0 18.5 261 18.7 18.2Berea Sandstone Samples

1-inch diameter 1.5 -inch diameterPorosity (%) Porosity (%)

std dev 24 0.5 0.4 22 0.4 0.1

Xmean 111 18.9 18.6 120 19.1 19.2std dev 24 0.8 0.6 22 0.8 0.4

Xmean 3.2 14.0 13.8 3 13.8 13.7std dev 0.9 0.6 0.5 0.7 0.7 0.7

Alundum Samples

Bedford Limestone Samples

Interlaboratory comparison - 25 labs (Sprunt et al , 1990)

Routine Porosity Distribution

Routine Porosity Histogram

0.14

0.16

0.18

atio

n

0.02

0.04

0.06

0.08

0.10

0.12

Frac

tion

of P

opul

a

0.000-2 2-4 4-6 6-8 8-10 10-12 12-14 14-16 16-18 18-20 20-22 22-24

Routine Helium Porosity (%)



Porosity Distribution by Basin

0 30

0.35

0.40

0.45at

ion

Al l BasinsGreater Green RiverWashakieUintaPiceanceWind River

0.05

0.10

0.15

0.20

0.25

0.30

Frac

tion

of P

opul

a

Powder River

• Distribution influenced by sampling – not normally distributed

0.000-2 2-4 4-6 6-8 8-10 10-12 12-14 14-16 16-18 18-20 20-22 22-24


Porosity Statistics by Basin All Greater Wind Powder

Basins Green Washakie Uinta Piceance River RiverRiver

Mean 7.1 7.3 9.5 6.1 6.1 5.8 13.2ea 3 9 5 6 6 5 8 3Median 6.2 4.6 8.7 5.9 6.1 5.5 15.1St Dev 5.1 6.4 5.4 4.2 3.8 3.3 4.5Minimum 0.0 0.0 0.0 0.0 0.0 0.0 2.6Maximum 24.9 23.6 23.8 22.2 24.9 13.2 16.9Kurtosis 0.7 -0.4 -0.4 1.1 4.5 -0.8 1.0Skewness 1.0 1.0 0.5 0.9 1.4 0.1 -1.5Count 2209 568 395 539 596 83 28



Statistics of Paired SamplesPorosity Histogram

0.400.450.50

atio

n

0.80.91.0

0.050.100.150.200.250.300.35

Frac

tion

of P

opul

a

0.10.20.30.40.50.60.7

• Histogram of ratio of paired plug porosities to mean porosity of plug pair. n = 652 x2= 1304

0.00

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

1.9

2.0

Paired Plugs Porosity Ratio

0.0

Grain Density Grain Density Histogram

0.25

0.30

latio

n

0.00

0.05

0.10

0.15

0.20

Frac

tion

of P

opul

• Mesaverde grain density is normally distributed for entire population (n=2200)

<2.56 2.56-2.58

2.58-2.60

2.60-2.62

2.62-2.64

2.64-266

2.66-2.68

2.68-2.70

2.70-2.72

> 2.72




Mesaverde Grain Density All Greater Wind Powder


Mean 2.653 2.648 2.660 2.639 2.660 2.673 2.679Median 2.654 2.645 2.662 2.649 2.661 2.673 2.674St D 0 040 0 029 0 034 0 052 0 038 0 029 0 026

• Statistically meaningful differences exist among

St Dev 0.040 0.029 0.034 0.052 0.038 0.029 0.026Minimum 2.30 2.50 2.47 2.30 2.35 2.51 2.60Maximum 2.84 2.77 2.79 2.80 2.84 2.73 2.75Kurtosis 15.1 2.6 3.7 13.2 14.0 10.2 3.9Skewness -2.00 0.28 -0.18 -2.82 -1.19 -1.87 -0.28Count 2184 566 393 532 583 82 28

basins• Low density minerals: carbonaceous fragments

(1.2-1.4 g/cc), K-feldspar (2.57 g/cc), Illite/smectite (2.60 g/cc)

Grain Density by Basin

Grain Density Histogram

0.50

0.60

latio

n All BasinsGreater Green RiverWashakie

0.00

0.10

0.20

0.30

0.40

Frac

tion

of P

opul Washakie

UintaPiceanceWind RiverPowder River

<2.56 2.56-2.58

2.58-2.60

2.60-2.62

2.62-2.64

2.64-266

2.66-2.68

2.68-2.70

2.70-2.72

> 2.72




Generic Porosity vs Confining Pressure

(after Byrnes, 1994)

Crack Compressibility• Crack porosity is far more

compressible than normal intergranular porosity

• Walsh & Grosenbaugh (1979) developed a model for fracture

ibili h h d llcompressibility that matches data well and can be expressed, as shown by Ostersen for low-k sandstones, by a linear porosity change with logarithmic change in stress

(after Walsh & Grosenba

(after Ostensen, 1983)

(after Walsh & Grosenba



Stress-Dependence of Porosity

0.9

1.0ro

sity

0.5

0.6

0.7

0.8

Frac

tion

of Ii

nitia

l Por

• Crossplot of fraction of initial pore volume versus net confining stress for 113 Mesaverde samples. Every sample exhibits a log-linear relationship though slopes and intercepts differ.

0.410 100 1000 10000

Net Confining Pressure (psi)

Pore Volume CompressibilityCformation = ΔVpore/Vpore

Δpσz

Stress field defined by σx, σy, σz

σhydro = K1σz – K2Pinital + K3 (Pinitial-P)Effective stress equation:

K1 = (σx+σy+σz)/3σz; lithostatic stressesK2 = (1-Cb/Cgr); Biot α – effect of pore pressure

σy

σx

Cf ti = K3 Ch d K2 (1 Cb/Cgr); Biot α effect of pore pressureK3 = K2 ((1+ν)/(3-3ν)); effect of pore pressure change, “uniaxial correction”; ν=Poisson’s ratio

Cformation K3 Chydro

Rock Type K1 K2 K3Consolidated Sandstone 0.85 0.80 0.45Friable Sandstone 0.90 0.90 0.60Unconsolidated Sandston 0.95 0.95 0.75Carbonate 0.85 0.85 0.55

(after Yale et al, 1993)



Type Compressibility CurvesUnconsolidated Friable Consolidated

A -0.00002805 0.0001054 -0.00002399B 300 500 300C 0.1395 -0.225 0.0623D 0.0001183 -0.00001103 0.00004308

Cf = A(σ-B)C + D

σ=K1Pover-K2Pi+K3(Pi-P)

30

40

50

60

me

Com

pres

sibi

lity

psi/1

0^6)

UnconsolidatedFriableConsolidated

0

10

20

0 2,000 4,000 6,000 8,000 10,000

Effective Lab Stress (psi)

Pore

Vol

um (

(after Yale et al, 1993)

Pore Volume Compressibility

-0.05

0.00

nge

Slo

pe

(

-0.25

-0.20

-0.15

-0.10

lativ

e Po

re V

olum

e C

han

1/ps

i)

• Crossplot of slope of log-linear curves in Figure 4.1.6 with porosity. • The relationship between the slope and porosity can be expressed: • Slope = -0.00549 -0.155/φ0.5

-0.300 2 4 6 8 10 12 14 16 18 20 22 24


Rel




1.25

1.30

1.35

Cha

nge

i)

1.05

1.10

1.15

1.20

1.25R

elat

ive

Pore

Vol

ume

CIn

terc

ept (

1/ps

i

• Crossplot of intercept of log-linear curves in Figure 4.1.6 with porosity. The relationship between the intercept and porosity can be expressed:

• Intercept = 0.013 φ + 1.08

1.000 2 4 6 8 10 12 14 16 18 20 22 24



• The above equations result in a power-law relationship between pore volume p pcompressibility and net effective confining pressure of a form:

log10 β = C log10 Pe + D• The slope and intercept of the pore volume

compressibility relations can be predicted using:C = -1.035 + 0.106/φ0.5

D = 4.857 φ-0.038



-0.98

-0.97

-0.96

-0.95

Com

pres

sibi

lity

-op

e (1

/psi

)

4.554.604.654.704.754.80

Com

pres

sibi

lity

-In

terc

ept

log10 β = C log10 Pe + D

-1.02

-1.01

-1.00

-0.99

0 5 10 15 20 25Routine Porosity (%)

log

Pore

Vol

ume

CPr

essu

re S

l

4.254.304.354.404.454.50

0 5 10 15 20 25Routine Porosity (%)

log

Pore

Vol

ume

Pres

sure

log10 β C log10 Pe + D• Where:

C = -1.035 + 0.106/φ0.5

D = 4.857 φ-0.038


1000

y (1

0^6/

psi)

10

100

re V

olum

e C

ompr

essi

bilit

y

φ = 21%φ = 18%φ = 15%φ = 12%φ = 8%φ = 6%φ = 4%φ 2%

β =10^[(-1.035+0.106/φ0.5)*log10 Pe+(4.857φ-0.038)]

1100 1000 10000

Net Effective Confining Stress (psi)

Por φ = 2%



In situ vs. Routine Porosity• φi/φo = A logPe + B

φi/φo Slope = A = -0.00549 – 0.155/φ0.5

φi/φo Intercept = B = 1.045 + 0.128/φ φi/φo Intercept B 1.045 0.128/φ

Where: φi = porosity at defined effective in situ stress Pe, φo = reference initial porosityPe = effective confining stressA and B are empirical constants that vary with rock

iproperties

In situ vs. Routine Porosity

1618202224

000

psi (

%)

Mesaverde StudyT ravis PeakMesaverde/FrontierClinton/MedinaLinear (Mesaverde Study)

02468

101214

0 2 4 6 8 10 12 14 16 18 20 22 24

Poro

sity

at P

e =

4,0

All Studies: φi = A φroutine + BMesaverde Study: φi = 0.96 φroutine – 0.73

Travis Peak: φi = 0.95 φroutine – 0.3Mesavrd/Frontier φi = 0.998 φroutine – 0.8Clinton/Medina: φi = 0.966 φroutine + 0.02

0 2 4 6 8 10 12 14 16 18 20 22 24

Routine Porosity (%)

Travis Mesaverde/ Clinton/ MesaverdePeak Frontier Medina Study

A > 0.950 0.998 0.966 0.960B > -0.300 -0.800 0.020 -0.734

Routine Porosity2.0 1.6 1.2 2.0 1.2

24.0 22.5 23.2 23.2 22.3




Porosity from Wireline Logs

• Densitye s y• Neutron• Sonic• NMR• NMR

PermeabilityPermeability



Core Analysis• Porosity & Grain Density

– Lithologic & other controls– Routine helium– In situ– Pore volume compressibility


f i l iPore volume compressibility• Permeability

– Routine Air– Klinkenberg– Crack & Capillary– Liquid– In situ– Effective & Relative


• Enhanced Oil Recovery– Chemical (polymer, surfactant, caustic)– Miscible (CO2, N2, Enriched Gas)– Thermal (Steam, Combustion)

• Electrical & Acoustic Properties• Gas-oil, Oil-water, Gas-water• Drainage, imbibition• Steady-state, unsteady-state• Single-phase stationary• Parameters influencing kr



• Electrical & Acoustic Properties – Archie Electrical Properties

• Cementation & Saturation Exponent, Cation Exchange

– Vp & Vs• Rock Mechanics

– Young’s Modulus, Poisson’s Ratio, Bulk Modulus

– Fracture Pressure

Original Darcy Flow Measurement

Q = k A dPµ dhµ dh

Analogs in Electric and heat flow

i = 1 A dVd dxd dx

dQ = KH A dTdx



Evolution of Permeability Modelingk=fr2/8

K=Φ/(FsAs2) x (L/La)2

(after CoreLab, 1978)(after Dullien, 1992)

Current Permeability Modeling• Permeability

controlled by:y– pore body size– pore throat size– distribution– connectivity– larger-scale

architecture



Comparison of Sandstone Pore Volume Distribution Measured by Hg Porosimetry

and Photomicroscopy

(after Dullien & Dhawan, 1974)

Liquid Permeability

Q = k A dPµ dL

(liquid)

Q = Volumetric Flow rate (cc/sec)K = Permeability (Darcies)A = Cross-sectional area (cm2)dP = Pressure differential (atm)m = fluid viscosity (centipoise)dL = Length (cm)

(after CoreLab, 1978)

Q = k A (P12-P2

2)µ 2PbzdL

(gas)



Permeability Definitions• Absolute Permeability (k) – Permeability of rock 100%

saturated with fluid of interest• Effective Permeability (keg, keo, kew) – Permeability to fluid

of interest when other fluids are also present in pore space• Relative Permeability (krg, kro, krw) – ke/k, Ratio of effective

to absolute permeability (reference for absolute may be effective at some condition, e.g. keo,Sw/keo,Swi)

• In situ – under reservoir conditions• Klinkenberg – Corrected for low pressure gas slippage

effects• Air – Permeability to air uncorrected for Klinkenberg y g

effect• Routine – Air permeability, generally measured with a

confining stress of less than ~500 psi

Permeability Determination• Full-diameter

– Influenced by microfractures– Averages response of

individual beds

• Probe mini-permeability– Fast– Allows high sampling

densityindividual beds– Possible drilling mud invasion– Less biased

• Plug– Precisely accurate– Possible sampling bias– May miss important beds

y– Accurate for k > 1md

• Chip– Low accuracy– Severe sampling bias

• Percussion Sidewall– Shattereday ss po ta t beds

• Drilled Sidewall– Greater sampling uncertainty– Similar to plug

– Under- and over-estimates properties

• Cuttings– Rarely used– Surface-to volume issues– Sever sampling bias



Klinkenberg Gas Slip

Gaskgas = kliq (1+4cl/r) = kliq (1+b/P)

measurable fluid velocity at wall

LiquidWhere;c = proportionality factor ~ 1l = mean free path at Pr = radius of capillaryb = proportionality constant

=f(r,l,kliq)P = pressure (atm)

Zero fluid velocity at wall

measurable fluid velocity at wall

10

100

or (p

si) Heid et al, 1950

Jones & Owens, 1981 - low k

p ( )Since b is a function of pore radius, mean free path at P, and liquid permeability it can vary from one low k sample to another but values are generally consistent with the Heid et al (1950) graph shown

(after Heid et al, 1950)

0.01

0.1

1

1E-04 0.001 0.01 0.1 1 10 100 1000Klinkenberg Permeability (md)

Klin

kenb

erg

b fa

ctb = 0.867 kliq

-0.33

b = 0.777 kliq-

0.39

General Correlation of Klinkenberg b Factor and Permeability

100

psi) Heid et al, 1950

Jones & O w ens 1981 - low k

0 .1

1

10

kenb

erg

b fa

ctor

( Jones & O w ens , 1981 - low k

b = 0 867 k -0.33

b = 0.777 kliq-0.39

0.011E-04 0 .001 0 .01 0 .1 1 10 100 1000

Klinkenberg Permeability (md)

Klin

k b = 0.867 kliq0.33



Correlation between Gas Slip-factor, b, and Permeability

(after Sampath & Keighin, 1982)

In situ Klinkenberg Permeability

10

100

1000

b fa

ctor

(atm

)

0.1

1

10

1E-08 1E-07 1E-06 1E-05 0.0001 0.001 0.01 0.1 1 10 100 1000


Klin

kenb

erg

b

k k (1 + 4 L/ ) k (1+b/P)

b = 0.851 kik-0.34 (Present Study)

b = 0.867 kliq-0.33 (Jones & Owens)

b = 0.777 kliq-0.39 (Heid)

kgas = kliquid (1 + 4cL/r) = kliquid (1+b/P) Gas

Liquid

kgas = gas permeability at pore pressurekliquid is liquid permeability and = Klinkenberg permeability kklinkc = proportionality constant (~ 1)L = mean free path of gas molecule at pore pressurer = pore radiusb = proportionality constant (=f(c, L, r))P = pore pressure (atm)



10

100

mea

bilit

ySandstone

Carbonate

Measured Insitu Klinkenberg vs Air Permeability

kik =0.685kia 1.12

0.01

0.1

1

u K

linke

nber

g Pe

rm(m

d)

R2 = 0.98

0.0001

0.001

0.0001 0.001 0.01 0.1 1 10 100In situ Air Permeability (md)

In s

itu

(after Byrnes, 2003)

Comparison of Klinkenberg Prediction Models

1

md)

Byrnes, 2003Jones & Owens, 1981

0.0001

0.001

0.01

0.1

kenb

erg

Perm

eabi

lity

(m

kklink = 0.685 kair1.12

0.000010.00001 0.0001 0.001 0.01 0.1 1

Air Permeability (md)

Klin

J&O (1980): kklink = 10^(-0.0398 logkair2+1.067logkair-0.0825)

valid for upstream pressure = 100 psi



Effect of Partial Water Saturation on Gas Slip

(after Sampath & Keighin, 1982)

“Averaging” Permeability Data• Permeability is a vector• Pseudo-Permeability is

direction dependent

• End-member models– Series Flow

direction dependent• Pseudo-Permeability

“averaging” is a function of flow model (3-D arrangement) assumed– Dependent on geomodel and

assumptions of smaller scale

– Parallel Flow– Random Flow– Vertical flow constraint

• Permeability is frequently scale dependentassumptions of smaller scale

permeability distribution dependent



Typical distributions of Porosity and Permeability

SeriesFlow

Permeability ArchitectureEnd Members

Parallel

HeterogeneousFlow

No vertical cross-flowVertical crossflow

kv=0, kv=Ckh

ParallelFlow



0.01 md 100 mdKarith = 1.010 mdKgeom = 0.011 md

Karith = 99.000 mdKgeom = 91.201 md

• In parallel flow the high perm drives the system• In series flow the low perm drives the system• Cross-flow influences parallel flow in closed systems (see Simulation Section)

0.01 md

100 md

100 md

0.01 md

Flow

100

ft

1 ft0.

01 m

d

0.01

md

100

md

100

md

100

md

0.01

md Kharm = 0.990 md

Kgeom = 91.201 mdKharm = 0.010 mdKgeom = 0.011 md

Core Plug Sampling with Bedding

C

BeddingPlanes

C - Suitable

A - Unsuitable

B – PossiblyB – Possibly suitable



Fraction of Upper Layer Thickness to Total hickness = 0.3

Model of Measured vs Composite Permeability for Layered Samples

Permeability-Porosity Equation : k = 3.65 x 10-5 e(0.68 Φ)

Upper Base Porosity Upper Base Average Permeability Measured RatioLayer Layer Difference Layer Layer Porosity for Average Permeability Measured/

Porosity Porosity Permeability Permeability Porosity Composite(%) (%) (%) (md) (md) (%) (md) (md) Permeability

0 14 14 0.0000365 0.497 9.8 0.0286 0.348 12.22 14 12 0.000142 0.497 10.4 0.0430 0.348 8.14 14 10 0.000554 0.497 11.0 0.0646 0.348 5.46 14 8 0.00216 0.497 11.6 0.0972 0.349 3.68 14 6 0.00841 0.497 12.2 0.146 0.350 2.4

10 14 4 0.0327 0.497 12.8 0.220 0.358 1.612 14 2 0.128 0.497 13.4 0.331 0.386 1.214 14 0 0.497 0.497 14.0 0.497 0.497 1.016 14 -2 1.94 0.497 14.6 0.747 0.929 1.218 14 -4 7.54 0.497 15.2 1.124 2.61 2.320 14 -6 29.4 0.497 15.8 1.690 9.16 5.421 14 -7 58.0 0.497 16.1 2.072 17.7 8.622 14 -8 114 0.497 16.4 2.541 34.7 13.623 14 -9 226 0.497 16.7 3.116 68.1 21.924 14 -10 446 0.497 17.0 3.821 134.1 35.1

Parallel Beds and Sampling• When sample contains

parallel beds of different k the measured k at the average porosity is

10

100

cula

ted

md) 30

35

40&

Upp

er

(%)

Measured Permeability - KmeasCalculated Permeability - KcalcRatio Kmeas/KcalcUpper Bed Porosity

average porosity is always greater than the k calculated for the composite of the individual beds

0.01

0.1

1

Mea

sure

d or

Cal

cPe

rmea

bilit

y (

0

5

10

15

20

25

Rat

io K

mea

s/K

calc

B

ed P

oros

ity

10

ed 35

40

pper

Measured Permeability - KmeasCalculated Permeability - KcalcRatio Kmeas/KcalcUpper Bed Porosity

9 10 11 12 13 14 15 16 17Average Porosity (%)

0.001

0.01

0.1

1

7 8 9 10 11 12 13Average Porosity (%)

Mea

sure

d or

Cal

cula

tePe

rmea

bilit

y (m

d)

0

5

10

15

20

25

30

Rat

io K

mea

s/K

calc

& U

pB

ed P

oros

ity (%

)

Upper Bed Porosity



General Lithologic Controls on the Effect of Overburden Pressure on Permeability

Effect of Confining Pressure on Permeablity

• Early work by Thomas and Ward (1972) Shows the 0.9

1.0

ility

Shows the characteristic decrease in permeability with increasing confining pressure exhibited by low-permeability sandstones

• Samples from Gas buggy well, Pictured 0 2

0.3

0.4

0.5

0.6

0.7

0.8

on o

f Ini

tial P

erm

eabi

buggy well, Pictured Cliffs Fm Rio Arriba Co., NM and Wagon Wheel well, Ft. Union Fm, Sublette Co., WY

0.0

0.1

0.2

0 1000 2000 3000 4000 5000 6000

Confining Pressure (psi)

Frac

ti



Effect of Confining Pressure on Spirit River and Cotton Valley Permeability

(after Walls, 1982)

Permeability Response to Confining Stress for Varying Crack Aspect Ratios

(after Brower & Morrow, 1983)

k/ki = {1-(16(1-n2)cLc)/(9(1-2n)pwi)s}3



Model Type Model Equation .Noncrack Capillary tube k/ki = (1-2s/E)4

Noncrack Gangi, grain, 1978 k/ki = {1-2{3p(1-n2)s/4E}2/3}4

Crack Jones &Owens, 1980 k/ki = {1-Slog(Pk/1000)}3

2 3

Models of Stress Dependent Permeability(after Ostensen, 1983)

Crack Brower & Morrow, 1983 k/ki = {1-(16(1-n2)cLc)/(9(1-2n)pwi)s}3

Asperity Gangi, bed of nails, 1978 k/ki = {1-(s/lE)e}3

Asperity Walsh, exp. dist., 1981 k = Ls3/12 {ln[(nE(prcs3)1/2)/(2(1-n2)s)]}3

Asperity Ostensen, Gauss.,1983 k = 0.76Ls3/12 {ln[(2.48E(s/rc)1/2)/(3p1.5(1-n2)s)]}2

Council Grove LimestonesMesaverde & Frontier

(after Jones &* Owens, 1980) (after Byrnes et al, 2001)

Sheet-like Pores in Travis Peak Sandstone

Transmitted light, 100X Fluorescent epoxy

8,275 ft, k = 0.007 md; SFE Well 2, Waskom Field, Harrison Co., TX(after Soeder & Chowdiah, 1990)



Pressure and Pore Throats

20

25en

cy

High P

Low P

5

10

15

e Si

ze F

requ

e(%

)Low P

0

5

0.01 0.1 1Pore Throat Diameter (um)

Pore

In situ vs Routine Permeability

10

100

eabi

lity Council Grove

Mesaverde/Frontier

0 001

0.01

0.1

1

Klin

kenb

erg

Perm

e(m

d)

logkik = 0.0588 (logkair)3

0.00001

0.0001

0.001

0.001 0.01 0.1 1 10 100Routine Air Permeability (md)

In s

itu K –0.187 (logkair)2

+1.154 logkair - 0.159



Known for many years that lowKnown for many years that low--K K sandstones are stress sensitivesandstones are stress sensitivey = -0.0088x3 - 0.0716x2 + 1.3661x - 0.4574

22

3

(mD

)

Stress dependence of permeability

sandstones are stress sensitivesandstones are stress sensitiveGeneralized = Generalized = f f (P(Pporepore, Lith), Lith)1997 Byrnes equation:1997 Byrnes equation:kkikik = 10^[1.34 (logk= 10^[1.34 (logkairair) ) -- 0.6] 0.6] This study:This study:kkikik = 10^[0.0088 (logk= 10^[0.0088 (logkairair))33 -- 0.072 0.072 (logk(logkairair))22+ 1.37 logk+ 1.37 logkairair +0.46]+0.46]

R2 = 0.9262

-6

-5

-4

-3

-2

-1

0

1

g In

situ

Klin

kenb

erg

Perm

eabi

lity

Statistically similar except for k > Statistically similar except for k > 1 mD1 mDno meaningful stress dependence no meaningful stress dependence over 10 mDover 10 mD

-7-7 -6 -5 -4 -3 -2 -1 0 1 2 3

log Routine Air Permeability Ppore = 100 psi (mD)

log

Permeablity Distribution

0.25

0.30

0.35

pula

tion

0.00

0.05

0.10

0.15

0.20

001-

001

001-

001

001-

001

001-

001

0.01

1-0.

1

0.1-

1

1-10

-100

1000

Frac

tion

of P

op

Distribution of in situ Klinkenberg permeability measured at 26.7 MPa (4,000 psi) net effective stress for all samples

0.00

000

0.00

00

0.00

000.

000

0.00

00.

00

0.00 0.0

0.00

1-0

0.01 0 1

10-

100-

1




In situ Klinkenberg Permeability Histogram

0 40

0.50

0.60pu

latio

n All BasinsGreater Green RiverWashakieUinta

0.00

0.10

0.20

0.30

0.4000

01-

0001

0001

-00

01

0001

-00

01

0001

-0.

001

-0.0

1

01-0

.1

0.1-

1

1-10

0-10

0

-100

0

Frac

tion

of P

op PiceanceWind RiverPowder River

Distribution of in situ Klinkenberg permeability measured at 26.7 MPa (4,000 psi) net effective stress by basin

0.00

000.

000

0.00

00.

00

0.00 0.0

0.0 0

0.00

1

0.0 10

100-


Permeability Statistics All Greater Wind Powder


Mean logk -2.60 -2.49 -2.03 -2.66 -2.95 -3.44 -1.88Median logk -2.93 -3.15 -2.46 -2.86 -3.03 -3.36 -2.21St Dev log 1.58 1.94 1.78 1.36 1.13 0.69 1.39Minimum logk -6.19 -6.19 -5.66 -5.33 -5.23 -5.11 -4.29Maximum logk 2.31 2.31 2.08 1.88 2.05 -1.98 0.55Kurtosis 0.62 -0.54 -0.39 0.17 4.02 -0.49 -0.38Skewness 1.05 0.79 0.76 0.74 1.48 -0.01 0.50Count 2143 555 373 529 577 81 28Mean 0.0025 0.0032 0.0094 0.0022 0.0011 0.0004 0.0133Median 0.0012 0.0007 0.0035 0.0014 0.0009 0.0004 0.0062St Dev 37.9 87.4 59.9 23.0 13.4 4.9 24.5Minimum 0.000001 0.000001 0.000002 0.000005 0.000006 0.000008 0.000051Maximum 206.0 206.0 121.0 76.2 112.2 0.010 3.53a u 06 0 06 0 0 6 0 0 0 3 53Kurtosis 0.62 -0.54 -0.39 0.17 4.02 -0.49 -0.38Skewness 1.05 0.79 0.76 0.74 1.48 -0.01 0.50Count 2143 555 373 529 577 81 28



Permeability Histogram

0 140.160.180.20

latio

n

0 70.80.91.0

0 000.020.040.060.080.100.120.14

Frac

tion

of P

opu

0 00.10.20.30.40.50.60.7

• Histogram of ratio of paired plug in situ Klinkenberg permeabilities to mean permeability of plug pair. n = 634 x2 = 1268

0.00

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

1.9

2.0

3.0

4.0

5.0 >6

Paired Plugs Permeability Ratio

0.0

Permeability vs Porosity

• Permeability a function of:• Permeability a function of:• Permeability a function of:Grain sizeShale bed architecturePore-throat sizePorosityDiagenetic alteration (including cementation)

• Permeability a function of:Grain sizeShale bed architecturePore-throat sizePorosityDiagenetic alteration (including cementation)g ( g )

• Porosity is optimal predictor parametric with lithofacies

g ( g )

• Porosity is optimal predictor parametric with lithofacies



Permeability as a Function of Grain Size and Sorting

(after Jonas & McBride, 1977)

Influence of Grain Size on Permeability

(from Shanley, 2004)



Permeability vs. Porosity by Grain Size

1

10

100

1000

000

psi,

mD

)

0.000001

0.00001

0.0001

0.001

0.01

0.1

Klin

kenb

erg

Perm

eabi

lity

(4,

X(4-9)XXX

X3XXX

X(0-2)XXX

• Generally subparallel trends increasing in porosity range and permeability at porosity with increasing grain size

• Influence of other variables significant

0.00000010 2 4 6 8 10 12 14 16 18 20 22 24

In situ calc Porosity (%)

K

Dispersed Clay Types in Sandstones Affecting Flow

Discrete ParticleKaolinite

Pore-LiningChlorite

Pore-BridgingIllite

(after Neasham, 1977)

Kaolinite ChloriteMontmorillonite

IlliteMixed-Layer



Influence of Clay types on Permeability

Discrete-particle, pore-lining and pore-bridging Kaolinite, Chlorite, and Illite can each result in permeability decrease by a factor of 1-0.03, 0.2-0.01, and 0.06-0.003, respectively

(after Wilson, 1981)

Discrete Particles-Pore Lining Kaolinite

American Hunter Old Road 8360’ (courtesy John Webb)



Pore Lining Clays

Chlorite

Mixed-Layer Illite-Smectite

American HunterOld Road 5490 ft

(courtesy John Webb)

Illite - Pore Bridging



Permeability vs Porosity• Generalized trend kik = 10[0.3φi-4.75] with 10X error• Different k-φ trends among basins due to lithologic variation• Beyond common k↑ with grain size↑, lithologic influence changes with porosity -

nonlinear

1000

0.001

0.01

0.1

1

10

100

rmea

bilit

y (4

,000

psi

, mD

)

Green RiverPiceancePowder River

0.0000001

0.000001

0.00001

0.0001

0 2 4 6 8 10 12 14 16 18 20 22 24In situ calc Porosity (%)

Klin

kenb

erg

Per

UintahWashakieWind RiverlogK=0.3Phi-3.7logK=0.3Phi-5.7

Permeability vs Porosity• logkik = 0.282φi + 0.182RC2-

5.13 (+4.5X MLRA)• logkik = 0.034φi

2-0.00109φi3 +

0.0032RC2 - 4.13 (+4.1X MNLRA)0.001

0.01

0.1

1

10

100

1000

erm

eabi

lity

(4,0

00 p

si, m

D)

X9XXXX8XXXX7XXXX6XXXX5XXXX4XXXX3XXXX2XXX

( )• Artificial Neural Network +3.3X

0.0001

0.001

0.01

0.1

1

10

100

1000

Pred

icte

d in

situ

Klin

kenb

erg

Perm

eabi

lity

(mD

)

10

100

1000

bilit

y (m

D)

0.0000001

0.000001

0.00001

0.0001


Klin

kenb

erg

P X2XXXX1XXX

0.000010.00001 0.0001 0.001 0.01 0.1 1 10 100 1000

Measured in situ Klinkenberg Permeability (mD)

hidden layer: 1Hidden layer nodes: 10

Mean> 8.239 4.280 6.294 hidden layer-Std Dev> 5.260 1.335 2.527 to-output

weightsNode Constant Phii RC2 RC4Constant -0.388

1 -0.760 2.946 -2.027 -6.438 -0.8852 -2.155 4.637 1.279 0.895 2.3233 -4.999 7.901 0.957 3.167 -2.5834 -1.484 -0.307 -1.695 6.175 -0.1545 -4.597 4.582 1.568 0.730 4.0226 -2.609 0.320 -2.201 -2.257 -2.4957 -1.765 -1.843 -1.122 0.145 -3.8598 2.839 -3.146 -9.237 0.264 0.7899 -1.566 1.029 -1.588 -3.390 2.400

10 2.951 0.778 3.316 0.179 -2.136

Input-to-hidden layer weights

0.0000001

0.000001

0.00001

0.0001

0.001

0.01

0.1

1

0 2 4 6 8 10 12 14 16 18 20 22 24Calculated in situ Porosity (%)

in s

itu K

linke

nber

g Pe

rmea

b

1XX9X1XX8X1XX7X1XX6X1XX5X1XX4X1XX3X1XX2X1XX1X1XX0X



Permeability vs Porosity•• Overall trend allows prediction of Overall trend allows prediction of KikKik from porosity with 10X errorfrom porosity with 10X error•• Multivariate linear equations using: 1) porosity, 2) rock class (1Multivariate linear equations using: 1) porosity, 2) rock class (1--3), and for each of three 3), and for each of three

porosity classes separately (0porosity classes separately (0--12%, 1212%, 12--18%, >18%), performed separately for each 18%, >18%), performed separately for each basin, exhibit an average standard error of prediction of: 0basin, exhibit an average standard error of prediction of: 0--12%: 3.612%: 3.6++2.4X; 122.4X; 12--18%: 18%: 3.33.3++3.6X; >18%: 3.1X (for all basins undifferentiated for this high porosity class); 3.6X; >18%: 3.1X (for all basins undifferentiated for this high porosity class); where the range of error for each standard error of prediction indicates the range of where the range of error for each standard error of prediction indicates the range of standard error among basinsstandard error among basinsstandard error among basinsstandard error among basins

•• Beyond common kBeyond common k↑ with grain size↑, ↑ with grain size↑, lithologiclithologic influence changes are complex and influence changes are complex and nonlinearnonlinear

0 01

0.1

1

10

100

1000bi

lity

(4,0

00 p

si, m

D)

Green River

0.0000001

0.000001

0.00001

0.0001

0.001

0.01


Klin

kenb

erg

Perm

eab Green River

PiceancePowder RiverUintahWashakieWind RiverlogK=0.3Phi-3.7logK=0.3Phi-5.7

Berea Cotton Valley Canyon

Chacra Wilcox Frontier-Moxa

Cleveland Travis ComparisonCleveland TravisPeak Comparison

of Tight Gas Sand k-fTrends

(from Dutton et al, 1993)



Generalized Tight Gas Sandstone Permeability vs Porosity Trends

10

100m

d) logki = 0.32+0.10 Φi - 5.05+1.48

0.01

0.1

1

itu P

erm

eabi

lity

(

BereaCotton ValleyCanyonFrontier-Moxa ArchWilcoxChacraClevelandTravis PeakMesaverde-GGRB

Data from various sources including Dutton et al, 1993; Byrnes, 2003; Castle and Byrnes, 2005)

0.0001

0.001

0 5 10 15 20 25In situ Porosity (%)

In s Medina

Mesaverde-Uinta

Stressed Permeability Hysteresis• Loading cycles approach similar values near original

reservoir stress• Successive loading cycles cease to exhibit furtherSuccessive loading cycles cease to exhibit further

hysteresis after second loading cycle

(after Thomas & Ward, 1968) (after Warpinski & Teufel, 1990)



Calculating Directional Permeability in Festoon

Cross-Bed Sets

(after Weber, 1982)

Shale Bed Continuity Distribution in Sandstone

Depositional Environmentsp

(after Weber, 1980)



Conclusions•• Grain density, porosity, and permeability measured on ~1500 Grain density, porosity, and permeability measured on ~1500

unique samples and 700 duplicates (5X original proposal)unique samples and 700 duplicates (5X original proposal)•• Core plugs obtained from 44 wells representing approximately Core plugs obtained from 44 wells representing approximately

7,000 feet of described core7,000 feet of described core•• Average grain density for 2200 samples is 2.654+0.033 g/cc Average grain density for 2200 samples is 2.654+0.033 g/cc

((±±1sd) 1sd) –– but grain density distributions differ slightly among basins & but grain density distributions differ slightly among basins &

lithofacieslithofacies. . •• Porosity variance with 1Porosity variance with 1--2 inches (2.52 inches (2.5--5 cm) = 5 cm) = ++10% (1sd)10% (1sd)•• Pore volume compressibility shows a logPore volume compressibility shows a log--linear relationship linear relationship

characteristic of sheet like pores and crackscharacteristic of sheet like pores and cracks

log10 β = C log10 Pe + D where C = -1.035 + 0.106/φ0.5 D = 4.857 φ-0.038

•• Lower porosity rocks exhibit greater pore volume compressibility Lower porosity rocks exhibit greater pore volume compressibility than high porosity rocks consistent with observed than high porosity rocks consistent with observed φφii vsvs φφroutineroutinetrendstrends

Conclusions•• KlinkenbergKlinkenberg slip term “b” consistent with prior trends to 1 slip term “b” consistent with prior trends to 1 μμDD•• Geometric mean permeability = 0.0025 Geometric mean permeability = 0.0025 mDmD, median = 0.0012 , median = 0.0012 mDmD•• Stress dependence of permeability is consistent with prior work Stress dependence of permeability is consistent with prior work

(Byrnes, 1997)(Byrnes, 1997)( y )( y )•• PorosityPorosity--permeability data exhibit two permeability data exhibit two subtrendssubtrends with with

permeability prediction approaching 5X within eachpermeability prediction approaching 5X within each–– Adding rock types or using an ANN model improves perm Adding rock types or using an ANN model improves perm

prediction to 3.3X prediction to 3.3X –– 4X4X•• Multivariate linear equations using: 1) porosity, 2) rock class (1Multivariate linear equations using: 1) porosity, 2) rock class (1--

3), and for each of three porosity classes separately (03), and for each of three porosity classes separately (0--12%, 1212%, 12--18%, >18%), performed separately for each basin, exhibit an 18%, >18%), performed separately for each basin, exhibit an average standard error of prediction of: 0average standard error of prediction of: 0--12%: 3.612%: 3.6++2.4X; 122.4X; 12--18%: 3.318%: 3.3++3.6X; >18%: 3.1X (for all basins undifferentiated for 3.6X; >18%: 3.1X (for all basins undifferentiated for this high porosity class); where the range of error for each this high porosity class); where the range of error for each standard error of prediction indicates the range of standard error standard error of prediction indicates the range of standard error among basinsamong basins


Byrnes: Capillary Pressure, Electrical Properties, Relative Permeability

Saturation &Saturation &Saturation & Capillary

Saturation & Capillary PressurePressure

Water Saturation• Water saturations in reservoir determined

using three basic methodsWi li l– Wireline logs

• Electric logs• NMR logs

– Fluid saturations from core• Routine core• Sponge corep g• High-pressure core• Oil- & low-invasion and water-based mud

– Capillary pressure measurements on core



Influence of Core Flushing with Water-based Mud on Saturations

(after CoreLab, 1982)

“Averaging” Saturation Data

• Saturation is a scalar Σ S φ •h

i=n

but is dimensionless• Sw should not be

averaged • BVW is averaged and

then converted back to Sw

Swaverage =

Σ φihii=1

i=n

Σ Swi φi •hii=1

Sw(Averaging for a well by thickness)



Buckles Plot – Piceance Basin

70

80

90

100ai

ton

(%) MWX-1

MWX-2MWX-3Buckles 600Buckles 300

20

30

40

50

60

70

outin

e C

ore

Wat

er S

atur

a

Buckles 240Buckles 180

Trendlines shown represent Sw = Aφ-1.1 where A = 180. 240. and 300, respectively. Differences in trends can be postulated to be due to differences in grainsize and/or clay type/content.

0

10

0 1 2 3 4 5 6 7 8 9 10 11 12 13Routine Core Porosity (%)

Ro

Buckles Plot – Piceance Basin

70

80

90

100

raito

n (%

) 4800-49355475-54855700-58456420-65557080-7180

10

20

30

40

50

60

70

Rou

tine

Cor

e W

ater

Sat

ur 7230-73607800-78908100-8120Buckles 7852-7863Buckles 7848-7877Buckle 7873-7886

• Routine core analysis porosity versus water saturation for the Piceance Basin MWX-2well. Saturation versus porosity trends exhibit commonly observed Buckles power-law relationship. Trendlines for depth intervals 7852-7886 shown represent Sw = Aφ-1.1 where A = 180. 240. and 300, respectively. Differences in trends can be postulated to be due to differences in grainsize and/or clay type/content.

0

10

0 1 2 3 4 5 6 7 8 9 10 11 12 13Routine Core Porosity (%)

R



Drop Cohesive Forces

Forceout = π r2 ΔPForce = 2π r σσ Forcein = 2π r σAt equilibrium:

Fout=Fin

π r2 ΔP = 2π r σrearranging

σ

σP1

rearrangingΔP = 2σ/r

Where :σ=interfacial tension (dyne/cm)r = radius (cm)

P2

Capillary Pressurerliq = rcap/cosθPc = Pnw Pwr Pc = Pnw-Pw

= 2σ/rliq

= 2σcosθ/rcap

rcap

rliqPnw

Pw q



Capillary Pressure in Uniformly Variable Capillary

(after Lake, 2005)• Pc = 2τ cosθ/rPc = capillary pressureτ = interfacial tensionθ = contact angler = pore radius

Capillary Rise

Pnw

Pnw

r

Pw

Pw

Pw

Pw* Pw* Pw*

h

hh

Pnw

FreeWaterLevelPnw=Pw

Pnw

Pw-Pnw = (ρw-ρnw) h g

Water



Capillary Pressure Equations• Pc = 2τ cosθ/r

where:P ill

• H = Pcres . (σbrine-σoil,gas) x 0.433

Where:H = height above free water levelPc = capillary pressure

τ = interfacial tensionθ = contact angler = pore radius

Pcres = Pcair-Hg τcosθres

τcosθair-hg

H height above free water levelPcres = reservoir capillary pressurePcair-Hg = air-mercury Pcσbrine = specific density of brine (g/cc)σoil,gas = specific density of oil or gas (g/cc)0.433 = conversion from density (g/cc)

to pressure gradient (psi/ft)

water

P

r = 2τ cosθ/PcH

rH

rB

oil

PhH

PwH

PhB

PwB

Capillary Pressure Equations• Pc = 2τ cosθ/r• r = 2τ cosθ/Pc

h

• H = Pcres .(σbrine-σoil,gas) x 0.433

• Pcres = Pcair Hg τcosθreswhere:Pc = capillary pressureτ = interfacial tensionθ = contact angler = pore radius

Pcres Pcair-Hg τcosθres

τcosθair-hg



Dep

th

(after Doveton, 1999)

Capillary Pressure MeasurementMercury Injection Porous Plate

Centrifuge

• Air-mercury• Air-brine• Oil-brine• Gas-oil

• Drainage• Imbibition



Mercury Capillary Pressure

(after Jennings, 1981)

Capilary Pressure Measurement

• Three different i H

high-P fluid

In situ Mercury Intrusion Unconfined (routine) Mercury Intrusion

air-Hg measurements– Unconfined

(n=150)– In situ

• Drainage-

Res

ista

nce

Ref

eren

ceC

ell

hi h P

Cor

e P

lug

hi h P

Cor

e Pl

ug

gimbibition (n=37)

• Drainage only (n=90)

• NES = 4000 psi

high -Pcore holder

mercury in

electricinsulator

Pressuretransducer

high -Pcore holder

mercury in

Pressuretransducer



Unconfined Capillary Pressure

78000

900010000

C ill P

30004000

50006000

7000

Air-Hg Capillary Pressure (psia)

• Capillary Pressure Varies with Lithofacies and associated pore size distribution and

0 10 20 30 40 50 60 70 80 90 100

01000

2000

Wetting Phase Saturation (%)

permeability

Capillary 1000

10000

ssur

e (p

sia)

Capillary Pressure Varies with Lithofacies and associated

pore size distribution

100

ercu

ry In

ject

ion

Pres 0.00025md

0.00049md0.0012md0.0017md0.0018md0.0030md0.0040md0.0057md0.0085md0.012md0.013md0.032md0.046md0.085md0.25md0.41md0.56md

100 10 20 30 40 50 60 70 80 90 100


Me 0.56md

0.84md2.24md



Normalizing Capillary Pressures• Capillary pressure curves change with permeability and

porosity• To predict water saturation from capillary pressure it is

necessary to either– Know the specific conditions at a given point and use a appropriate

measured capillary pressure– Construct a synthetic capillary pressure curve for the conditions at the

point– Develop a relation between a normalized capillary pressure function

and saturation• Two principal approaches for normalization or synthetic curve

iconstruction:– Leverett “J” function (Leverett, 1941)– Unpublished normalization of Brooks-Corey l function (Brooks and

Corey, 1964)• Fractal model extension of B-C

Leverett J function• J(Sw) = CPc (k/φ)0.5/τcosθ

– J = dimensionless Pc function, function of SwSw

– C = conversion constant = 0.2166– Pc = capillary pressure (psi)

– τcosθ = interfacial tension (dyne/cm) X cosine of the contact angle (degrees)

– k = permeability (md)

– φ = porosity (fraction)



Basic Leverett J Function• At its simplest a Leverett J

function is constructed by plotting taking a series of capillary pressure curves forcapillary pressure curves for samples of different porosity and permeability and plotting the J value versus the water saturation

• From the cross-plot a curve is constructed that honors th d tthe data

• For some formationsSw = -Alog10J + B

• Valid J<1; For J>1 then Sw=Swi

•A problem with the Leverett J function is the wide variance in saturation that occurs near the “irreducible” water saturation which is the saturation of principal interest for many analyses

Leverett J Adjustment for Swi• Because of the problem that Leverett J functions can have

near the “irreducible” water saturation (Swi) aspects of the Brooks-Corey method have been adopted to improve the J-S l ti b li i f S iSw correlation by normalizing for Swi

• Water saturation is normalized using:– Swe = (Sw-Swi)/(1-Swi) where Swe = effective water saturation,

Sw = water saturation at any given Pc and Swi = “irreducible” water saturation

– Method is dependent on criteria for defining Swi• Plot of J versus log Swe is generally linear with a constant

slope, λ, and an intercept, J*, related to the J function normalized threshold entry pressure.

• The calculation of water saturation requires knowledge or back-calculation of Swi:

J = J* Swe(1/λ)



Normalization: Leverett J Function• J function

works poorly for mixed 7

8

90.00025md0.00049md0.0012md0 0017mdfor mixed

lithofacies and between basins

• Does work OK for single lithofacies in a small area

3

4

5

6

7

Leve

rett

J Fu

nctio

n

0.0017md0.0018md0.0030md0.0040md0.0057md0.0085md0.012md0.013md0.032md0.046md0.085md0 25mdsmall area

0

1

2

0 10 20 30 40 50 60 70 80 90 100


0.25md0.41md0.56md0.84md2.24md

Normalized Brooks-Corey• Brooks and Corey (1966) showed that a log-log

plot of Pc versus Swe often exhibits a linear trend with slope λ and intercept equal to the thresholdwith slope, λ, and intercept equal to the threshold entry pressure

• logSwe = -λlogPc + λlogPce for Pc>Pce– Pc=capillary pressure– Pce = threshold entry pressure– Swe = (Sw-Swi)/(1-Swi)– λ = slope of log-log plot

Capillary pressure parameters, λ and Pce, are correlated with permeability and/or porosity to develop Pc curves



Normalization: Brooks-Corey Capillary Pressure

• Transform taking logarithm of Pc and Sw• λ represents pore throat size distribution• Standard unimodal curves can be reduced

to intercept (Pce = extrapolated threshold 2000

3000

4000

5000

6000

7000

8000

9000

10000

Hg

Cap

illar

y Pr

essu

re (p

sia)

entry) and slope (λ)0

1000

2000

0 10 20 30 40 50 60 70 80 90 100Wetting Phase Saturation (%)

Air

-

10000

ssur

e (p

sia)

Pc = 1.54E+07Sw-2.05

R2 = 0.997

1000

10000

y Pr

essu

re (p

sia)

Pceλ

100

1000

0 10 20 30 40 50 60 70 80 90 100


Air

-Hg

Cap

illar

y Pr

es

100

1000

10 100


Air

-Hg

Cap

illar

y

Change in Methane Density with Pressure and Temperature

0.300.32

ρ=0.03861-0.0003331T+5.943*10-5P-4.287*10-9P2+1.226*10-13P3

0 080.100.120.140.160.180.200.220.240.260.28

hane

Den

sity

(g/c

c) Pressure (psia)12000

11000

10000

9000

8000

7000

6000

5000

4000

3000

0.000.020.040.060.08

90 100 110 120 130 140 150 160 170 180 190 200 210 220 230

Temperature (deg F)

Met

h 3000

2000

1000



Brine Density vs P-T-XBw = (1 +ΔVwp)x(1-ΔVwT); Bw =FVFw

γw = 1+ 6.95x10-6 XTDS; γw = specific gravity, X mg/l

ΔVwT = -1.0001x10-2 + 1.339x10-4T+5.5065x10-7T2

ΔVwp = -1.953x10-9pT-1.7283x10-13p2T-3.5892x10-7p-2.2534x10-10p2

ρw = γw/Bw

1.05

1.10

1.15

1.20

1.25

1.30

ensi

ty (g

/cc)

65 F, 15 psi

65 F 1000 psi

65 F, 5000 psi

65 F, 10000 psi

100 F, 15 psi

100 F, 1000 psi

100 F, 5000 psi

100 F, 10000 psi

200 F, 15 psi

200 F, 1000 psi

200 F, 5000 psi

0.90

0.95

1.00

0 50 100 150 200 250 300

Total Dissolved Solids (mg/l/1000)

De , p

200 F, 10000 psi

300 F, 15 psi

300 F, 1000 psi

300 F, 5000 psi

300 F, 10000 psi

Discrepancy in High P,T Methane-Water Interfacial Tension

• IFT data of Hough, Raza, and Wood (1951) hibi IFT 70

80

(1951) exhibits IFT <30 dyne/cm at higher P,T

• Data of Jennings & Newman (1971) exhibit higher values

• J&N data more30

40

50

60

70

Mod

eled

IFT

(dyn

e/cm

)

J&N data more consistent, HRW may have had unknown problem with system elastomer seal contamination

0

10

20

0 10 20 30 40 50 60 70 80HRW Measured IFT (dyne/cm)

J&N

M



Relationship Between Pore Throat Diameter and Permeability by Lithology

0 445

100

y = 2.61x0.445

R² = 0.9259

0.1

1

10

ore

Thro

at D

iam

eter

(μm

)

Ss lithic

0.001

0.01

0.000001 0.00001 0.0001 0.001 0.01 0.1 1 10 100 1000 10000

Prin

cipa

l Po

Insitu Klinkenberg Permeability (md)

Ss lithicSs arkosicSs quartzoseLs interparticleLs chalkLs moldicLs oomoldic


100

)

Dp = 7.17(k/φ)0.49

R2 = 0.83

0.1

1

10

l Por

e Th

roat

Dia

met

er (μ

m)

Lithology

0.001

0.01

0.000001 0.00001 0.0001 0.001 0.01 0.1 1 10 100 1000

Prin

cipa

Porosity Normalized Permeability (kik/φa, md/%)

LithologySs lithicSs arkosicSs quartzoseLs interparticleLs chalkLs moldicLs oomoldic




100

m)

Dp = 7.17(k/φ)0.49

R2 = 0.83

0.1

1

10

pal P

ore

Thro

at D

iam

eter

(μm

Lithology

Ss lithic

Ss arkosic

Ss quartzose

Ls interparticle

Ls chalk

Ls moldic

0.001

0.01

0.000001 0.00001 0.0001 0.001 0.01 0.1 1 10 100 1000

Prin

cip

Porosity Normalized Permeability (kik/φa, md/%)

Ls oomoldic

Mesaverde Hi

Mesaverde Lo

Power (Lithology)

Relationship Between Threshold Entry Pressure and Permeability

10000

ry kak

kmk

10

100

1000

rcur

y Th

resh

old

Entr

Pres

sure

(psi

)

kmkkik

y = 64.66x-0.44

R2 = 0.82

1

10

1E-06 0.00001 0.0001 0.001 0.01 0.1 1 10 100

Klinkenberg Permeability (mD)

Air-

Me



Stress effect on Pc

1

10

100

1000

10000


Air

-Hg

Cap

illar

yPr

essu

re(p

sia)

R091255.9 ftk = 113 mD

= 24.5%φ

1000

10000

ure

(psi

a)

1

10

100

1000

10000


Air-

Hg

Cap

illar

yP

ress

ure

(psi

a)

R7802729.9 ftk = 7.96 mD

= 19.2%φ

1000

10000

ure

(psi

a)

113 mD 8 mD

• no significant difference in high-low pairs at high K

1

10

100


Air-

Hg

Cap

illar

yPr

essu

LD43C4013.25 ftk = 0.190 mD

= 12.9%φ

10

100

1000

10000

Air

-Hg

Cap

illar

yPr

essu

re(p

sia)

10

100

1000

10000

r-Hg

Cap

illar

yP

ress

ure

(psi

a)

1

10

100


Air

-Hg

Cap

illar

yPr

essu

E9466486.4 ftk = 0.637 mD

= 12.2%φ 0.6 mD 0.2 mD

• increasing Pce separation with decreasing K

• merging of curves at 35-50% Sw• smaller pores are in

1

10

100

1000

10000


Air

-Hg

Capi

llary

Pres

sure

(psi

a)

PA4244606.5 ftk = 0.00107 mD

= 12.7%φ

1

10

100

1000

10000


Air

-Hg

Capi

llary

Pres

sure

(psi

a)

B02913672.5 ftk = 0.000065 mD

= 2.6%φ

10 10 20 30 40 50 60 70 80 90 100


A

B02911460.6 ftk = 0.0255 mD

= 4.4%φ

10 10 20 30 40 50 60 70 80 90 100


Ai

E9466530.3 ftk = 0.0416 mD

= 9.5%φ 0.04 mD 0.02 mD

0.001 mD 0.00007 mD

protected pore space

• users of Winland R35 need to adjust for confining stress

• threshold entry pressure is predictable from √K/φ at any

y = 11.77x0.50

R2 = 0.77

y = 11.28x0.50

R2 = 0.930 1

1

10

100

old

Entr

yPo

reD

iam

eter

( μm

)

√K/φ at any confining pressure

• correct unconfined Pce to insitu Pce

0.01

0.1

1E-06 0.00001 0.0001 0.001 0.01 0.1 1 10 100

Klinkenberg Permeability/Porosity (mD/%)

Thre

sho

A

y = 6 75x-0.50

1000

10000

sC

olum

n) to insitu Pce

based on perm change with stress

y = 6.48x-0.50

R2 = 0.77

y = 6.75xR2 = 0.93

1

10

100

1E-06 1E-05 0.0001 0.001 0.01 0.1 1 10 100

Klinkenberg Permeability/Porosity (mD/%)

Thre

shol

dEn

try

Ga

Hei

ght(

ft)

C



Stress effect on Pc

1

10

100

1000

10000

Air

-Hg

Cap

illar

yPr

essu

re(p

sia)

R091 1

10

100

1000

10000

Air-

Hg

Cap

illar

yP

ress

ure

(psi

a)

R780

y = 11.77x0.50

R2 = 0.77

y = 11.28x0.50

R2 = 0.93

0.01

0.1

1

10

100

1E-06 0.00001 0.0001 0.001 0.01 0.1 1 10 100

Thre

shol

dEn

try

Pore

Dia

met

er( μ

m)

A10 10 20 30 40 50 60 70 80 90 100


R091255.9 ftk = 113 mD

= 24.5%φ

1

10

100

1000

10000


Air-

Hg

Cap

illar

yPr

essu

re(p

sia)

LD43C4013.25 ftk = 0.190 mD

= 12.9%φ

10 10 20 30 40 50 60 70 80 90 100


R7802729.9 ftk = 7.96 mD

= 19.2%φ

100

1000

10000

llary

Pres

sure

(psi

a)

100

1000

10000

lary

Pre

ssur

e(p

sia)

1

10

100

1000

10000


Air

-Hg

Cap

illar

yPr

essu

re(p

sia)

E9466486.4 ftk = 0.637 mD

= 12.2%φ

113 mD 8 mD

0.6 mD 0.2 mD

Klinkenberg Permeability/Porosity (mD/%)A

y = 6.48x-0.50

R2 = 0.77

y = 6.75x-0.50

R2 = 0.93

1

10

100

1000

10000

1E 06 1E 05 0 0001 0 001 0 01 0 1 1 10 100

Thre

shol

dEn

try

Gas

Col

umn

Hei

ght(

ft)

•• threshold entry pressure threshold entry pressure is entirely predictable is entirely predictable from √K/from √K/φφ ratio at any Pratio at any P

1

10

100

1000

10000


Air

-Hg

Capi

llary

Pres

sure

(psi

a)

PA4244606.5 ftk = 0.00107 mD

= 12.7%φ

1

10

100

1000

10000


Air

-Hg

Capi

llary

Pres

sure

(psi

a)

B02913672.5 ftk = 0.000065 mD

= 2.6%φ

1

10


Air

-Hg

Cap

i

B02911460.6 ftk = 0.0255 mD

= 4.4%φ

1

10


Air-

Hg

Cap

il

E9466530.3 ftk = 0.0416 mD

= 9.5%φ 0.04 mD 0.02 mD

0.001 mD 0.00007 mD

1E-06 1E-05 0.0001 0.001 0.01 0.1 1 10 100

Klinkenberg Permeability/Porosity (mD/%)C

Brooks-Corey Slope• PSD expressed by Pcslope• Pcslope = f (k)• Pcslope ↓ with P ↑

Leverett J(Sw) = Pc (k/φ)0.5/τcosθ

Implicitly assumesPcslope = Constant

Poor fit becausePcslope ≠ C = f(k, lith)

y = -0.037Ln(x) + 1.256R2 = 0.052

y = -0.0304Ln(x) + 1.87R2 = 0.0216

2

3

4

5

Cor

ey C

apill

ary

sure

Slo

pe

in situunconfined

0

1

2

1E-05 0.0001 0.001 0.01 0.1 1 10 100 1000In situ Klinkenberg Permeability (mD)

Bro

oks-

CPr

es



Modeled Pc Curves

Modeled Pc curves

400500600700800900

1000

abov

e fre

e w

ater

(ft) k=0.0001 mD

k=0.001 mDk=0.01 mD

k=0.1 mD

k=1 mD

k=10 mD

0100200300

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0Water Saturation (fraction)

Hei

ght a

Modeled Pc curves

100

1000

ree

wat

er (f

t)

Pc properties evolve

1

10

0.0 0.1 1.0Water Saturation (fraction)

Hei

ght a

bove

fr k=0.0001 mDk=0.001 mD

k=0.01 mD

k=0.1 mDk=1 mD

k=10 mD

Pc properties evolve over time as diagenesis changes porosity and pore architecture

Hysteresis of Capillary Pressure

Drainage-ImbibitionCycles

3

4

5

• Non-wetting residual saturation to imbibition S f(S i)

12

5

Snwr = f(Snwi)

(after Larson & Morrow, 1981)

Midale Dolφ = 23%



Capillary Pressure Hysteresis in Coarse Sand Pack

(after Klute, 1967)

Drainge-Imbibition

10000r

• what is the residual trapped gas when a reservoir leaks or along a gas migration path?

10

100

1000

0000

ght a

bove

Fre

e W

ater

Leve

l (ft)

Primary DrainageFirst ImbibitionSecondary DrainageSecond ImbibitionTertiary DrainageThird Imbibition

0.1

1

0 10 20 30 40 50 60 70 80 90 100


App

rox.

Hei



Capillary Pressure

Hysteresis1

10

100

1000

10000

0 10 20 30 40 50 60 70 80 90 100


Air-

Hg

Cap

illar

yP

ress

ure

(psi

a)


E393 7001.1ft = 17.4% = 28.9 mD

φ

kik

1

10

100

1000

10000


Air-

Hg

Cap

illar

yP

ress

ure

(psi

a)

Primary DrainagePrimary ImbibitionSecond DrainageSecond ImbibitionThird DrainageThird Imbibition

B049 9072.1 ft (A) = 12.3% = 6.74 mD

φ

kik

1000

10000

sure

(psi

a)


kik

1000

10000

sure

(psi

a)


kikkikkikkikkik

•• Composite primary Composite primary drainage trend drainage trend consistent with consistent with singlesingle--cycle drainagecycle drainage

•• ImibitionImibition curves curves hibi hi h ihibi hi h i

1

10

100


Air-

HgC

apill

ary

Pres

s

E393 7027.2 ft = 15.0% = 1.93 mD

φ

1

10

100


Air

-Hg

Cap

illar

yP

ress

R829 5618.3 ft (B) = 9.2% = 0.287 mD

φ

10

100

1000

10000

r-H

gC

apill

ary

Pre

ssur

e(p

sia)


10

100

1000

10000r-

Hg

Cap

illar

yP

ress

ure

(psi

a)


exhibit high trappingexhibit high trapping•• Trapped saturation Trapped saturation

increases with increases with increasing initial increasing initial saturationsaturation

10 10 20 30 40 50 60 70 80 90 100


Air

B646 8294.4 ft (B) = 7.6% = 0.022 mD

φ

10 10 20 30 40 50 60 70 80 90 100


Air

S685 6991.2 ft (B) = 8.6% = 0.0063 mD

φ

1

10

100

1000

10000


Air-

HgC

apill

ary

Pres

sure

(psi

a)


E458 6404.8 ft (A) = 9.5% = 0.0019 mD

φ

1

10

100

1000

10000


Air

-Hg

Cap

illar

yPr

essu

re(p

sia)


KM360 8185.7 ft (B) = 5.9% = 0.00070 mD

φ

Trapping increases with increasing initial saturation

(after Lake 2005)



Residual Non-wetting Phase Saturation

0.9

1.0

Snw

r)

0.3

0.4

0.5

0.6

0.7

0.8

Non

wet

ting

Phas

e Sa

tura

tion

(S

0.0

0.1

0.2

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Initial Nonwetting Phase Saturation (Snwi)

Res

idua

l N

Residual Gas SaturationC = 1/[(Snwr-Swi)-1/(Snwi-Swi)]Snwr = 1/[C + 1/Snwi]C = 0.55 (min ε); Swi = 0

C = 1/[(Snwr-Swi)-1/(Snwi-Swi)]Snwr = 1/[C + 1/Snwi]C = 0.55 (min ε); Swi = 0

Sample Swirr Land C C Land C Snwr SnwrCondition definition Average Standard Minimum Standard Std Error

Error Error Error C=0.55all Swirr = 1-Snwmax 0.57 0.329 0.53 0.077 0.077unconfined Swirr = 1-Snwmax 0.61 0.294 0.59 0.087 0.088hysteresis Swirr = 1-Snwmax 0.61 0.383 0.51 0.056 0.057confined Swirr = 1-Snwmax 0.44 0.249 0.45 0.088 0.085all Swirr = 0 0.73 0.443 0.63 0.073 0.073unconfined Swirr = 0 0.78 0.360 0.71 0.080 0.081hysteresis Swirr = 0 0.75 0.562 0.59 0.057 0.057confined Swirr = 0 0.61 0.316 0.54 0.078 0.078all Swirr = 0, Snwi<70% 0.70 0.054 0.053

0.5

0.6

0.7

0.8

0.9

1.0

g Ph

ase

Satu

ratio

n (S

nwr) unconfined Snwi= 1-Snwmax

unconfined hysteresisLand C =0.59, Swirr=0Land C=0.71, Swirr=0Land C =0.55, Swirr=0

unconfined Swirr = 0, Snwi<70% 0.83 0.062 0.061hysteresis Swirr = 0, Snwi<70% 0.70 0.052 0.051confined Swirr = 0, Snwi<70% 0.50 0.038 0.039

0.0

0.1

0.2

0.3

0.4

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0


Res

idua

l Non

wet

ting



Residual Saturation



0.9

1.0

wr) unconfined

confined

0.4

0.5

0.6

0.7

0.8

nwet

ting

Phas

e Sa

tura

tion

(Sn confined

Land C=0.66, Swi=0Land C =0.54, Swi=0

0.0

0.1

0.2

0.3

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0


Res

idua

l Non

Residual Gas Saturation

1

10

100

1000

10000

0 10 20 30 40 50 60 70 80 90 100


Air-

Hg

Cap

illar

yP

ress

ure

(psi

a)


E393 7001.1ft = 17.4% = 28.9 mD

φ

kik

1

10

100

1000

10000


Air-

Hg

Cap

illar

yP

ress

ure

(psi

a)


B049 9072.1 ft (A) = 12.3% = 6.74 mD

φ

kik

1000

10000

re(p

sia)


kik

1000

10000

re(p

sia)


kikkikkikkikkik

• Snwi and Snwr are ~ = for Sw > 80%• e g for Swi of 30% Swr is ~50%

• Snwi and Snwr are ~ = for Sw > 80%• e g for Swi of 30% Swr is ~50%

0.6

0.7

0.8

0.9

1.0

ase

Satu

ratio

n (S

nwr) unconfined

confinedLand C=0.66, Swi=0Land C =0.54, Swi=0

1

10

100


Air-

HgC

apill

ary

Pres

sur Third Imbibition

E393 7027.2 ft = 15.0% = 1.93 mD

φ

1

10

100


Air

-Hg

Cap

illar

yP

ress

ur Third Imbibition

R829 5618.3 ft (B) = 9.2% = 0.287 mD

φ

10

100

1000

10000

gC

apill

ary

Pre

ssur

e(p

sia)


10

100

1000

10000

gC

apill

ary

Pre

ssur

e(p

sia)


• e.g., for Swi of 30%, Swr is ~50%• e.g., for Swi of 30%, Swr is ~50%

0.0

0.1

0.2

0.3

0.4

0.5

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0Initial Nonwetting Phase Saturation (Snwi)

Res

idua

l Non

wet

ting

Ph

10 10 20 30 40 50 60 70 80 90 100


Air-

Hg

B646 8294.4 ft (B) = 7.6% = 0.022 mD

φ

10 10 20 30 40 50 60 70 80 90 100


Air-

Hg

S685 6991.2 ft (B) = 8.6% = 0.0063 mD

φ

1

10

100

1000

10000


Air-

HgC

apill

ary

Pres

sure

(psi

a)


E458 6404.8 ft (A) = 9.5% = 0.0019 mD

φ

1

10

100

1000

10000


Air

-Hg

Cap

illar

yPr

essu

re(p

sia)


KM360 8185.7 ft (B) = 5.9% = 0.00070 mD

φ



Residual gas

saturation0 7

0.8

0.9

1.0

tion

(Snw

r)

Complete trapping, C=0Vuggy, isolated moldic, C=0.3Mesaverde high C =0.35Mesaverde Ss, C=0.55Mesaverde low, C=0.9Cemented Ss, C=0.7Berea, C=1.7Unconsolidated sucrosic oolitic C=3

• Trapping constant, C consistent with cemented

0.3

0.4

0.5

0.6

0.7

ual N

onw

ettin

g Ph

ase

Satu

rat Unconsolidated, sucrosic, oolitic, C 3

cemented sandstone

0.0

0.1

0.2

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1Initial Nonwetting Phase Saturation (Snwi)

Res

idu

Electrical PropertiesElectrical PropertiesPropertiesProperties



Wireline log analysis tools

• Lithofacies identification

Permeability - 1Core

0.001 1000unknTimur : Constant Exponent

0.001 1000MDTimur : Variable Exponent

0.001 1000md

1:240 MD

in F

Reservoir ComponentsPorosity

0.4 0V/VPHIX

0.4 0.0V/VOil

0.4 0V/VWater

0.4 0V/VShale

0 2V/V

CPHI

0 0unknWater

Oil

Gas

0 1

Permeability - 2Core

0.001 1000unknTimur : Sw-Sw(Density)

0.001 1000unknTimur : Sw/Sw(Density)

0.001 1000unkn

6400

• Accurate porosity calculation

• Water saturation calculations

06425

64506475

65006525

65506550

MWX2

Resistivity of a simple rock model with straight pores

0 1Porosity0 1Porosity (Φ) RwResistivity

(Ro)The ‘formation factor’ (F) is defined as the ratio Ro/Rw

F = 1/φ

8



For a rock with a tortuous pore network ....

F = 1/φmF = 1/φThis is the first Archie equation, where ‘m’ is known as the ‘cementation exponent’

The resistivity of hydrocarbon-bearing rocks

01 Water saturation (Sw)Ro Resistivity (Rt)

The ‘resistivity index’ (I) is defined as the ratio Rt/Ro

I = 1/Sw

Water saturation (Sw)

n

8



Putting it all together ...the Archie equation

ΦF m=a

=oR

R 1ΦF mRwRR Swo

tI == n

1

φSw= a

m*wRt

1/n( R )

Core measurement of the formation factor, F

core plug

A

L

ro Rw

Φ

p g



When F and φ are plotted on logarithmic graph paper ...

1

Φ m= 3

0.1

F

m= 2

m= 1

1 10 100 10000.01

Regional Water Chemistry Database

DOE Contract DE-FC-02NT41437Billingsley et al

Advanced Resources International

Hi t i l D t• Historical Data• 3200 Well Locations

–Greater Green River Basin and Wind River Basin• 8000 Chemical Analyses• Access/Excel Formats



m in sandstones Archie (1942) observed the range in value of m in sandstones:1.3 unconsolidated sandstones1.4 - 1.5 very slightly cemented 1.6 - 1.7 slightly cemented1.8 - 1.9 moderately cemented2.0 - 2.2 highly cementedg y

Guyod gave the name “cementation exponent” to m, but noted that the pore geometry controls on m were more complex and went beyond simple cementation

m variability

Core measurements of formation factor andformation factor and porosity in a Cherokee sandstone sample, with a computed value of cementation exponent mfor each core sample from:from:

F = 1/φm



30

35

40on

(%)

Archie Cementation Exponents

Mesaverde Frontier

10

15

20

25

erce

nt o

f Pop

ulat

io

Medina

Mesaverde-Frontier

0

5

1.3-

1.4

1.4-

1.5

1.5-

1.6

1.6-

1.7

1.7-

1.8

1.8-

1.9

1.9-

2.0

2.0-

2.1

2.1-

2.2

Archie Cementation Exponent (m, a=1)

P

Water Saturation Calculations

• Archie• Simandoux• Fertl• Dual-Water• Waxman-Smits



Simandoux• Developed theoretically primarily for

Gulf Coast application

Whereφ = effective porosity

• Rw = water resistivity• Rt = formation true resistivity• Rsh = shale or clay resistivity• Vsh = volume of shale

Fertl• Developed for shaly sandstones in Rocky Mountains

Whereφ = effective porosity

• Rw = water resistivity• Rt = formation true resistivity• Vsh = volume of shale• A = Constant



Dual-Water/Waxman-Smits• (Clavier, Coats, and Dumanoir, 1984)

Swt - SwbS =

Whereφt = total porosity

1 - SwbSw =

• Rwf = formation water resistivity• Rt = formation true resistivity• Rwb = bound water resistivity (Rwa in shales)• Swt = total water saturation• Swb = bound water saturation (various methods for determination

– e.g., Swb = α vq Qv; vq = 0.28 cc/meq25oC, α(XNaCl) ≈ 1

Clay Surface Area & Cation Exchange Capacity

Clay Type Cation Exchange MorphologyPure Clay Clay in Sandstone Capacity (Meq/100g)

Specific Surface

Kaolinite 15-18 0.05-0.20 3-15 BooksFans

Smectite 85-100 0.5-2.0 80-150 Honeycomb

Illite 90-115 1.5-10 10-40 Curled flakes with projecting and fibrous mat

Smectite-Illite

85-115 0.5-10 10-150 Similar to Smectite Illite

(mixed-layer)

& Illite

Chlorite 40-60 0.5-2.0 10-40 Cardhouse, rosette(after Grim, 1968; Gaida et al, 1973)



Waxman and Smits (1969) Calculated Water SaturationWaxman and Smits (1969)

Calculated Water Saturation• Redefined the Archie equation including the• Redefined the Archie equation including the• Redefined the Archie equation including the

influence of conductive claysCo = (1/F*) (Cw + BQv)

• Co = core conductivity at Sw=100% (mho/m)Cw = water conductivity (mho/m)F* = salinity/clay conductivity independent formation factor

• Redefined the Archie equation including the influence of conductive clays

Co = (1/F*) (Cw + BQv)• Co = core conductivity at Sw=100% (mho/m)

Cw = water conductivity (mho/m)F* = salinity/clay conductivity independent formation factorQv = cation exchange capacity of the core (meq/cc)B = specific counter-ion activity [(equiv/l)/(ohm-m)]

• F*/F = (1 + BQv/Cw)

Qv = cation exchange capacity of the core (meq/cc)B = specific counter-ion activity [(equiv/l)/(ohm-m)]

• F*/F = (1 + BQv/Cw)

Waxman-SmitsWater Saturation Calculations

Waxman-SmitsWater Saturation Calculations

• Sw = [(F*Rw) Rt(1+ RwBQv/Sw)]1/n*

• F* = salinity/clay conductivity independent formation factorQv = cation exchange capacity of the core (meq/cc)B = specific counter-ion activity [(equiv/l)/(ohm-m)]

• Qv ≈ CEC(1-φ)ρma/100φ

• Sw = [(F*Rw) Rt(1+ RwBQv/Sw)]1/n*

• F* = salinity/clay conductivity independent formation factorQv = cation exchange capacity of the core (meq/cc)B = specific counter-ion activity [(equiv/l)/(ohm-m)]

• Qv ≈ CEC(1-φ)ρma/100φQv ( φ)ρma φQv ( φ)ρma φ



Waxman-Smits-Thomas

*

w *m*a

=w

RS

⎞⎛

φ*n

w

vw t

w

SBQR1R

⎟⎟⎟

⎠

⎞

⎜⎜⎜

⎝

⎛+

F* = a*/φm* Intrinsic Formation Factor; free of excess conductivity

m* Intrinsic cementation exponent; free of excess conductivityp ; y

n* Intrinsic saturation exponent; free of excess conductivity

Rw Resistivity of brine at temperature (ohm-m)

B Equivalent counterion conductance at temperature (1/ohm-m)/(equiv / liter)

Qv Cation exchange capacity per ml pore space (meq/ml)

Qv Lab Methods• Wet Chemistry

– Utilizes crushed rock with high surface area– Requires sample porosity & grain density toRequires sample porosity & grain density to

compute Qv– Crushing can improperly exposes Qv sites not

present in native pores• Multiple Salinity (Co vs Cw)

Fl th h f lti l li it b i– Flow-through of multiple salinity brines on core– Preserves distribution of clays and Qv – time – intensive



Multiple-salinity Analysis

*FQ B

*FCC vw

O +=

,1/R

oor

eC

ondu

ctiv

ity(C

O)

Clay-rich sandstone

*FQ B vmax

Slope @ Bmax brines = 1/F*

Excess conductivityC

Clean sandstone

Brine Conductivity (CW), 1/Rw0

BmaxQv

FCC

F1C W

WO =⋅=

Porosity dependence of “m”Empirical: m = 0.234 ln Empirical: m = 0.234 ln φφ + 1.33+ 1.33Dual porosity: m = log[(Dual porosity: m = log[(φφ--φφ22))m1m1 + + φφ22

m2m2]/log ]/log φφφφ22 = 0.35% m= 0.35% m11=2, m=2, m22=1; SE both = 0.11=1; SE both = 0.11rock behaves like a mixture of matrix porosity and rock behaves like a mixture of matrix porosity and p yp ycracks or fracturescracks or fractures

both models fit databoth models fit data

φφ = bulk porosity= bulk porosityφφ22 = fracture porosity= fracture porosity

1 5

1.6

1.7

1.8

1.9

2.0

2.1

2.2

Cem

enta

iton

Expo

nent

rin

e=40

Kpp

mN

aCl)

mm11 = matrix = matrix cementation cementation exponentexponent

mm22 = fracture = fracture cementation cementation exponentexponent

1.0

1.1

1.2

1.3

1.4

1.5

0 2 4 6 8 10 12 14 16 18 20 22


In s

itu A

rchi

e C

(m, a

=1, X

br



Archie Cementation Exponent

2.4)

• Empirical: m = 0.95 - 9.2φ + 6.35φ0.5

• Dual porosity: m = log[(φ-φ2)m1 + φ2m2]/logφ

1 5

1.61.7

1.81.92.0

2.1

2.22.3

ntat

ion

Expo

nent

(m,A

=1

• φ = bulk porosity• φ2 = fracture or

touching vug porosity High: m1 = 2 1 φ2 = 0 0005High: m1 = 2 1 φ2 = 0 0005

1.0

1.11.21.3

1.41.5

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 0.22 0.24Porosity (fraction)

Arc

hie

Cem

eporosity• m1 = matrix

cementation exponent

• m2 = fracture or touching vug cementation exponent

High: m1 2.1, φ2 0.0005Int: m1 = 2.0, φ2 = 0.001Low: m1 = 1.8, φ2 = 0.002

m2 = 1

High: m1 2.1, φ2 0.0005Int: m1 = 2.0, φ2 = 0.001Low: m1 = 1.8, φ2 = 0.002

m2 = 1

Archie porosity (cementation) exponentNearly all cores exhibit some salinity dependenceNearly all cores exhibit some salinity dependencetested plugs with 20K, 40K, 80K, and 200K tested plugs with 20K, 40K, 80K, and 200K ppmppm brinesbrines

1.0

0.4

0.5

0.6

0.7

0.8

0.9

Con

duct

ivity

(mho

/m) n=335

0.0

0.1

0.2

0.3

0 2 4 6 8 10 12 14 16 18 20 22

Brine Conductivity (mho/m)

Cor

e C



Archie Cementation Exponent vs. Rw

Nearly all cores Nearly all cores exhibit some salinity exhibit some salinity dependencedependencetested plugs with tested plugs with 2.0

2.1

2.2

2.3xp

onen

t,

p gp g20K, 40K, 80K, and 20K, 40K, 80K, and 200K 200K ppmppm brinesbrines

1.4

1.5

1.6

1.7

1.8

1.9

rchi

e C

emen

tatio

n Ex

(m, A

=1)

1.0

1.1

1.2

1.3

0.01 0.1 1

Brine Resistivity (ohm-m)

In s

itu A

r

Multi-salinity Archie m

2.0

2.2

2.4

onen

t (m

,

1.2

1.4

1.6

1.8

e C

emen

taito

n Ex

poa=

1)

200K

80K

40K

• Archie m decreases with decreasing salinity

0.8

1.0

0 2 4 6 8 10 12 14 16 18 20 22

Arc

hie


40K

20K



Slopem-logRwvs Porosity

0.0

0.1

0.2

w S

lope

Each core exhibits a highly linear m vs logRw

Mean value for all cores:

y = 0.0118x - 0.3551R2 = 0.1198

-0.8

-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0 2 4 6 8 10 12 14 16 18 20 22

In s

itu A

rchi

e m

vs

log

Rw Mean value for all cores:

Average Slopem-Rw = -0.27+0.32 (2 standard deviations)

where Slopem-Rw = slope of mRw versus logRw.

Slopes exhibit a weak correlation with porosity . This correlation

0 2 4 6 8 10 12 14 16 18 20 22In situ Porosity (%) can be used to improve the

prediction of m at any salinity:Slopem-Rw = 0.00118 φ – 0.355 (φ - %).

Estimation of Archie m• Each core exhibits a highly linear m vs logRw• Mean value for all cores:

– Average Slopem-Rw = -0.27+0.32 (2 standard deviations)h Sl l f l R– where Slopem-Rw = slope of mRw versus logRw.

• Slopes exhibit a weak correlation with porosity . This correlation can be used to improve the prediction of m at any salinity:– Slopem-Rw = 0.00118 φ – 0.355 (φ - %).

• Combining the above equations the Archie cementation exponent at any given porosity and reservoir brine salinity can be predicted using:using:– mX = m40 + Slopem-Rw (log RwX + logRw40K)– mX = (0.676 logφ + 1.22) + (0.0118 φ-0.355) x (logRwX + 0.758); φ<14%– mX = 1.95 + (0.0118 φ-0.355) x (logRwX + 0.758); φ>14%

• where mx = m at salinity X• m40 = m at 40K ppm NaCl, log RwX = log10 of resistivity of brine at salinity X• logRw40K = log10 of resistivity of 40K ppm NaCl = 0.758



Salinity dependence of “m”•• m = a ln m = a ln φ φ + b+ b•• a, b = a, b = ff (salinity)(salinity)

20K ppm

y = 0.2267Ln(x) + 2.2979

R2 = 0.6619

2 00

2.50

•• low porosity rocks hold low porosity rocks hold more gas than we more gas than we thoughtthought0.00

0.50

1.00

1.50

2.00

0.000 0.050 0.100 0.150 0.200 0.250

insitu porosity (%)

Axis Title

Series1

Log. (Series1)

40K ppm

y = 0.2328Ln(x) + 2.409

R2 = 0.6547

0.50

1.00

1.50

2.00

2.50

3.00

Axis Title

Series1

Log. (Series1) 80K ppm

y = 0.2149Ln(x) + 2.4354

R2 = 0.51322.50

3.00

0.00

0.000 0.050 0.100 0.150 0.200 0.250

insitu porosity (%)

0.00

0.50

1.00

1.50

2.00

0.000 0.050 0.100 0.150 0.200 0.250

insitu porosity (%)

Axis Title

Series1

Log. (Series1)

200K ppm

y = 0.1621Ln(x) + 2.3222

R2 = 0.3633

0.00

0.50

1.00

1.50

2.00

2.50

3.00

0.000 0.050 0.100 0.150 0.200 0.250

insitu porosity (%)

Axis Title

Series1

Log. (Series1)

Critical GasCritical GasCritical Gas Saturation

Critical Gas Saturation



Overview• Previous work indicated

that krg could be modeled using: Corey eqn with p=1.7 & Sgc ~

Western Sandstones

0.1

1

as R

elat

ive

Perm

eabi

lity

q p gc0.15-0.05*log10kik

• Swc ~ Swi600

• Issues– little krg data at Sw >

65% : Does p vary or

0.010 10 20 30 40 50 60 70 80 90 100

Water Saturation

Ga

0.1

1

bilit

y (f

ract

ion)

g-10 mdw -10 mdg-1 mdw -1 mdg-0.1 mdw -0.1 mdg-0.01 mdw -0.01 mdp y

Sgc vary or both?– little Swc data: how is

Swc = f (kik)? Or what is krw exponent ?

SgcSwc>Swi

0.0001

0.001

0.01

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Water Saturation (fraction)

Rel

ativ

e Pe

rmea g-0.001 md

w -0.001 md

Relative Permeability and Capillary Pressure

• Krg - relative permeability to gas• Krw - relative permeability to water• Sgc - critical gas saturation (Sg

f ti th) ativ

e Pe

rmea

bilit

y 1

krg krw

necessary for connective gas path)• Swc - critical water saturation (Sw

below which water relative permeability is zero or less than measurable threshold

• Swi - “irreducible” water saturation (Sw at which further increase in Pc, hydrocarbon column height, results in S d l th it i ry

Pre

ssur

eR

ela

t abo

ve F

ree

ter L

evel

0 10

SgcSwcSwi

Pcdrainage

curve

Transition

gas-

only

prod

uctio

nat

ertio

n

Sw decrease less than some criteria

• At Sg<Sgc no gas flow only water flow• At Swc<Sw<(1-Sgc) transition zone -

both gas & water flow• At Sw<Swc no measurable water flow

only gas flow

Water Saturation

Cap

illar

Hei

ght

Wa

0 10

zone

Free water levelwat

er-o

nly

prod

uctio

nga

s&w

apr

oduc



Relative Permeability Scaling

0.70.80.91.0

eabi

lity

0.01

0.1

1

mea

bilit

yLogarithmic Linear

A t ti h th iti l t ti f h h th

0.00.10.20.30.40.50.6

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0Water Saturation

Rel

ativ

e Pe

rm

0.000001

0.00001

0.0001

0.001

0 0

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Water Saturation

Rel

ativ

e Pe

rm

• As saturations approach the critical saturation for each phase the relative permeability for that phase changes by orders of magnitude

• At saturations above critical saturations the relative permeability to the remaining flowing phase changes less than an order of magnitude

Relative Permeability Reference Frame• krg = kreg/kr?• Relative permeability is the ratio of the effective

permeability of one phase to a baseline bili di i l fpermeability - traditional references are:

– kr = ke/kabsolute; where kabs may be kair,kwater, koil kklink– kr = ke/kenw,Swc or kr = ke/kenw,Swi

• kabs is the absolute permeability– In high k rocks kwater ~ kklink ~ kabs (~ kair)– In high k rocks ken S c ~ kabs and S c~S iIn high k rocks kenw,Swc kabs and Swc Swi– In low k rocks kwater<kklink– In low k rocks keg,Swi < kklink

• User must choose reference frame - (carefully)



Relative Permeability Reference Frame• Selection of

– kreference = kwater – kref = keg,Swc – results in krg > 1 at Sw < Swc

• For most reservoir simulation programs 0.001

0.01

0.1

1

10

ve P

erm

eabi

lity

kr cannot exceed 1• In reservoir Swi can be < Swc but it

achieved the low Sw by water flow at krw << krw,Swc

0.01

0.1

1

mea

bilit

y

0.000001

0.00001

0.0001

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Water Saturation

Rel

ativ

0.1

1

10

eabi

lity

kref = kwater

0.000001

0.00001

0.0001

0.001

0 0

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Water Saturation

Rel

ativ

e Pe

rm

0.000001

0.00001

0.0001

0.001

0.01

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Water Saturation

Rel

ativ

e Pe

rme

kref = kklink kref = keg,Swc

Generalized Drainage & Imbibition Relative Permeability Curves

Generalize Drainage Curves Generalized Imbibition CurvesGeneralize Drainage Curves Generalized Imbibition Curves

(after Sahimi, 1994)



Gas Relative Permeability of Low-Permeability Tight Gas Sandstone• Referenced to kklink

• Measurements performed at Sw<Swi by evaporation

• Note shift to lower krg at a given Swwith decreasing ki

(after Thomas & Ward, 1972)

Effect of Confining Pressure on Relative Permeability for Tight Gas Sandstone

• Note data points showing little effect of significant change in g gconfining pressure on krg

• Ward & Morrow (1987) data indicate that krg under pressure may be 10% less than at low pressure

• Referenced to kklink,P• Measurements• Measurements

performed at Sw <Swiby evaporation

• Note shift to lower krgat a given Sw with decreasing ki

(after Thomas & Ward, 1972)



Gas Relative Permeability is Similar using

different techniques to obtain water

saturation

(after Walls, 1982)

Influence of Confining Pressure on Gas Permeability with Core at Different Water Saturations

(after Walls, 1982)



Effect of Confining Pressure on Relative Permeability for Tight Gas Sandstone

Data of Randolph (1983) show moderate effect of confining stress on kr at low water saturation but increasing effect with increasing Sw

(after Randolph, 1983)

Single-phase Stationary krg Curves

• Relative gas permeability data, representing k values obtained at several saturationskrg values obtained at several saturations, were compiled from published studies (Thomas and Ward, 1972; Byrnes et al , 1979; Sampath and Keighin, 1981; Walls, 1981; Randolph, 1983; Ward and Morrow, 1987)



Gas Relative PermeabilityWestern Sandstones1

bilit

y Bounding curves consistent with

0.1

Rel

ativ

e Pe

rmea

b single-point data

0.010 10 20 30 40 50 60 70 80 90 100

Water Saturation

Gas

n=43

Single-point krg,Swi Data• Relative gas permeability data representing k• Relative gas permeability data, representing krg

values obtained at a single Sw and krg values obtained for a single sample at several saturations, were compiled from published studies (Thomas and Ward, 1972; Byrnes et al , 1979; Jones and Owens, 1981; Sampath and Keighin, 1981; Walls, p g1981; Randolph, 1983; Ward and Morrow, 1987; Byrnes, 1997; Castle and Byrnes, 1997; Byrnes and Castle, 2001)



Single-Sw Gas Relative PermeabilityAll Tight Gas Sandstones

0 80.91.0

bilit

y 1-10 md0.1-1 md0.05-0.1 md0 01 0 05 d

0.30.40.5

0.60.70.8

Rel

ativ

e Pe

rmea

b 0.01-0.05 md0.005-0.01 md0.001-0.005 md0.0001-0.001 md1 md0.1 md0.01 md0.001 md0.0001 md

0.00.10.2

0 10 20 30 40 50 60 70 80 90 100Water Saturation (%)

Gas

R

Relative Permeability to Gas Relative Permeability to Gas –– at Stressat StressMultiple reservoir intervals Multiple reservoir intervals –– GGRB (n = 583)GGRB (n = 583)

1.01.0

Krg/4000

Byrnesdata

Rel

ativ

e Pe

rmea

bilit

yR

elat

ive

Perm

eabi

lity

0.40.4

0.60.6

0.80.8

Water Saturation (%)Water Saturation (%)00 20201010 3030 4040 5050 6060 7070 8080 9090 100100

00

0.20.2

(Shanley et al, 2003)



Single-Sw Gas Relative PermeabilityAll Tight Gas Sandstones

1bi

lity

0.01

0.1

Rel

ativ

e Pe

rmea

b

1-10 md0.1-1 md0.05-0.1 md0.01-0.05 md0.005-0.01 md0.001-0.005 md0.0001-0.001 md

0.0010 10 20 30 40 50 60 70 80 90 100

Water Saturation (%)

Gas

R 1 md0.1 md0.01 md0.001 md0.0001 md

Relative Permeability Modeling• Early workers (e.g., Burdine, 1953) modeled kr based on Kozeny-

Carmen equation and capillary pressure curves and associated pore size distribution where kr was expressed as a function of the fraction of pore space occupied and the relative size occupiedp p p p

• Example: Wyllie & Spangler (1958)

krw = [(Sw-Swc)/(1-Swc)]2∫∫

0

Sw

1

dSw

Pc2

0dSw

Pc2Tortuosity Term

Gates and Lietz (1950)

Mean HydraulicRadius TermBurdine (1953)

∫

∫Sw

1

1

dSw

Pc2

0dSw

Pc2

krg = [1-(Sw-Swc)/(1-Sgc-Swc)]2

Gates and Lietz (1950)



Corey (1954) Equation• Corey (1954) making the approximation that 1/Pc

2

= C(Sw-Swc)/(1-Swc), i.e., is linear with ΔSw over a range of saturations, simplified the Burdine-P ll d i t ti tPurcell drainage type equations to:

S -S( )4

k

krg = Sw-Swc

1-Sgc-Swc(1- )2 Sw-Swc

1-Swc( )21-( )

Sw Swc

1-Swc( )krw =

•Exponents often modified to adjust for different pore size distribution

Key Features of Krg

Swc decreases with decreasing kiSwc

Sw-Swc,g

1-Sgc-Swc,g

Sw-Swc,gkrg = (1- )1.7

1-Swc,g( )21-( ) Swc,g = 0.16 + 0.053*log10kik

(where<0 then 0)

Scg = 0.15 - 0.05*log10kik

All Tight Gas Sandstones

0.01

0.1

1

Rel

ativ

e Pe

rmea

bilit

y

1-10 md0.1-1 md0.05-0.1 md0.01-0.05 md0.005-0.01 md0.001-0.005 md0.0001-0.001 md1 d

krg, at any given Swincreases with increasing ki

krg,Sw

0.0010 10 20 30 40 50 60 70 80 90 100


Gas

1 md0.1 md0.01 md0.001 md0.0001 md

Sgc increases with decreasing ki

Sgc

Krg curve shapes are approximately identical for widely different lithofacies



Key Features of Gas Relative Permeability in Low Permeability RocksSwc,g decreases with decreasing ki

Swc,gAll Tight Gas Sandstones

0.01

0.1

1

Rel

ativ

e Pe

rmea

bilit

y

1-10 md0.1-1 md0.05-0.1 md0.01-0.05 md0.005-0.01 md0.001-0.005 md0.0001-0.001 md1 d

krg, at any given Swincreases with increasing ki

krg,Sw

0.0010 10 20 30 40 50 60 70 80 90 100


Gas

1 md0.1 md0.01 md0.001 md0.0001 md

Sgc increases with decreasing ki

Sgc

Krg curve shapes are approximately identical for widely different lithofacies

Why is Sgc Important?

1

0.001

0.01

0.1

Rel

ativ

e Pe

rmea

bilit

y P = 1.7Sgc = f (kik)

P=f (kik)Sgc = 10%

0.00001

0.0001

0 10 20 30 40 50 60 70 80 90 100Water Saturation

Gas



Definitions• Critical-gas saturation has been defined variously as

– minimum gas saturation at which the gas phase flows freely (Firoozabadi et al., 1989)

– maximum gas saturation before any gas flow occurs (Moulomaximum gas saturation before any gas flow occurs (Mouloand Longeron, 1989)

– gas saturation at which gas freely flows to the top of a reservoir (Kortekaas and Poelgeest, 1989)

– gas saturation at which gas is produced at the outlet of a core (Li and Yortsos, 1991)Li d Y (1993) i l l ifi d b– Li and Yortsos (1993) appropriately clarified a robust definition as the gas saturation at which the gas forms a system-spanning cluster (and consequently flows freely). This definition is consistent with the critical percolation threshold at which the gas is connected to all parts of the system and not just flowing in a subset of the system.

Measured Sgc• 0.006 < Sgc < 0.38

– Solution-gas laboratory-measured (Hunt and Berry, 1956; Handy, 1958; Moulu and Longeron, 1989; Kortekaas and Poelgeest, 1989; Firoozabadi et al., 1989; and Kamath and Boyer, 1993)

• 0.03 < Sgc < 0.11– 0.0008 mD < kik < 0.031 mD, n =11, Chowdiah (1987)

• Sgc=0.01– k = 0.10 mD, Colton sandstone sample, Kamath and Boyer (1993)

• Sgc = 0.10– solution gas drive, k = 0.10 mD, Colton sandstone sample, Kamath and Boyer

(1993)

• Sgc=0.02– Torpedo sandstone, k = 413 mD, Closmann (1987)

• 0.045 < Sgc < 0.17– Schowalter (1979) , n=10, 0.01 mD < k < 30.09 mD



Published Single-Saturation Gas Relative Permeablity

0 10000

1.00000y

0.00100

0.01000

0.10000

as R

ealti

ve P

erm

eabi

lity

Thomas & Ward, 1972Byrnes et al, 1979Jones & Owens, 1980Sampath & Keighin, 1981Walls, 1981Chowdiah, 1990

0.00001

0.00010

0 10 20 30 40 50 60 70 80 90 100Water Saturation (%)

G Morrow et al, 1991Byrnes, 1992Byrnes, 1997Byrnes & Castle, 2000

Measurement of Snwc (Sgc)• Confined mercury intrusion

with electrical conductivity• Advantages

Hg in ΔV

– Percolation threshold of Hg detected by resistivity drop of >200x105 to <5 ohm

– Able to determine Pc equilibrium saturation after non-equilibrium breakthrough

Cor

e

q g– Determine pore throat size

difference between entry threshold and percolation threshold

High P Vesseloil

Pnetconfining = 4,000 psi



Critical Non-wetting Phase Saturation

0.16

0.18

0.20

0.22

g Ph

ase MICP-inflection

Electrical Resistance

0 02

0.04

0.06

0.08

0.10

0.12

0.14

Crit

ical

Non

-wet

ting

Satu

ratio

n

• Electrical conductivity and Pc inflection indicate 0% < Snwc < 22%• Higher Snwc in complex bedding lithofacies

0.00

0.02

0.00001 0.0001 0.001 0.01 0.1 1 10 100 1000


Measurement of Snwc (Sgc)• Confined gas injection• Advantages

– Sample water wetl i f fi b bbl i

N2 inmicropipette

gas bubble

– Expulsion of first gas bubble is highly sensitive

– Sgc from both Vgas and weight change

• Disadvantages– Potential saturation gradient

Cor

e

g– Solution gas development at

high pressure– Pore volume change with stress

and possible hysteresisHigh P Vesseloil

Pconfining = 4,000 psi




0.70.80.91.0

35404550

Sgc Histogram

0.00.10.20.30.40.50.6

05

1015202530

00 04 08 12 16 20 24 28 32 36 40 44 48

Freq

uenc

y

• Sgcavg = 0.066+0.13 (2 stdev)• Wide variance

0.0

0.0

0.0

0. 0. 0.2

0.2

0.2

0.3

0.3

0.4

0.4

0.4



0.35

0.40

0.45

0.50

urat

ion

0 00

0.05

0.10

0.15

0.20

0.25

0.30

Crit

ical

Gas

Sat

u

• Sgc is low for high permeability samples and fraction of population shows increasing Sgc with decreasing permeability

0.000.0001 0.001 0.01 0.1 1 10 100




How does Sgc get so high?• In cross-bedded sandstone

series intrusion requires Pc=threshold of lowest k 80

100

120

140

Pres

sure

(psi

)

0.1 md0.01 md0.001 md

1Pc=threshold of lowest k facies

• Sgc = f(Pc1&Pc2, V1/V2, Sgc1&Sgc2, Pc equilibrium, architecture)

0

20

40

60

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Water Saturation (fraction)

Gas

-Wat

er C

apill

ary

P

12

2

12

Sgc=75% Sgc=75%

Sgc=5%

Sgc=5%

1

12

Sgc vs beddingbedding

Corey and Rathjens(1956)



Sgc and PercolationInvasion direction

3) Series network N ( ) - preferential sample-

• Sgc (L) = A LD−E(Wilkinson and Willemsen, 1983)

– L is network dimension– A is a numerical constant (for

simple cubic network A = 0.65)– D is the mass fractal dimension

1) Percolation Network N ( ) - Macroscopically homogeneous, random distribution of bond sizes, e.g., Simple Cubic Network (z=6)

p

2) Parallel Network N ( ) preferential orientation of pore sizes or beds of different

II

Nspanning orientation of pore sizes or beds of different networks perpendicular to the invasion direction.

p

4) Discontinuous series network N ( ) - preferential non-sample-spanning orientation

d

800

900

1000

re(k

Pa)

0.001 md0.1 md

of the percolation cluster – E is the Euclidean dimension

• As L → ∞ Sgc → 0 – Sgc = 21.5% for L = 10– Sgc = 2.4% for L = 1000– Sgc = 0.8% for L = 10000)

• Experimental results can be explained using four - pore network architecture models

Np

networks parallel to the invasion direction.

p Np

N N

p p p gof pore sizes or beds of different networks perpendicular to the invasion direction. Represents continuum between and p.

0

100

200

300

400

500

600

700


Gas

-Wat

erC

apill

ary

Pres

sur

AB

Sgc andpercolation theory

•• critical gas saturation critical gas saturation strongly controlled by strongly controlled by sedimentary structures/rock sedimentary structures/rock f b if b i

Invasion direction

3) Series network N( ) - preferential sample- fabricfabric•• anyany bedding parallel bedding parallel

laminations result in low laminations result in low SgcSgc

1) Percolation Network N ( ) - Macroscopically homogeneous, random distribution of bond sizes, e.g., Simple Cubic Network (z=6)

p

2) Parallel Network N

N

( ) preferential orientation of pore sizes or beds of different

networks parallel to the invasion

II

3) Series network N

N

( ) - preferential sample-spanning orientation of pore sizes or beds of different networks perpendicular to the invasion direction.

p

4) Discontinuous series network N

Np

( ) - preferential non-sample-spanning orientation of pore sizes or beds of different networks

d

700

800

900

1000

sure

(kPa

)

0.001 md0.1 md

•• experimental results can be experimental results can be explained using four explained using four -- pore pore network architecture modelsnetwork architecture models

N networks parallel to the invasion direction.

p Np

N N

of pore sizes or beds of different networks perpendicular to the invasion direction. Represents continuum between and p.

0

100

200

300

400

500

600

700


Gas

-Wat

erC

apill

ary

Pres

s

AB



Prediction of Sgc• Four pore network architecture models:

– percolation (Np)– parallel (N//)– series (N⊥)

discontinuous series (N )– discontinuous series (N⊥d)

• Analysis suggests that Sgc is scale- and bedding-architecture dependent in cores and in the field.

• Sgc is likely to be very low in cores with laminae and laminated reservoirs (N//)) and low (e.g., Sgc < 0.03-0.07 at core scale and Sgc < 0.02 at reservoir scale) in massive-bedded sandstones of any permeability (Np)

• In cross-bedded lithologies exhibiting series network properties (N⊥), Sgcapproaches a constant reflecting the capillary pressure property differences and

l ti l th b d i i F th t k Srelative pore volumes among the beds in series. For these networks Sgc can range widely but can reach high values (e.g., Sgc < 0.6)

• Discontinuous series networks, representing lithologies exhibiting series network properties but for which the restrictive beds are not sample-spanning (N⊥d), exhibit Sgc intermediate between Np and N⊥ networks.

CMG IMEXSingle 1-ft thick High-Permeability Layered

Reservoir Simulation Model

• 1ft – 0.01, 0.1, 1, 10, 100 md• keg=0.004,0.04,0.4,4,40 md• Swc= 0.34, krg = 0.38• kbase= 0.004 md, kvert = 0.0004md



Base Model – keg=0.004 md

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+06

1.E+07

1.E+08

1.E+09

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

khigh = 4 md, kbase = 0.004 md

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)

1.E+07

1.E+08

1.E+09

1.E+10

J-01 J-02 J-03 J-04 J-05 J-06Time (m-yr)

Cum

ulat

ive

Gas

(scf

)



Effect of high-k thin-bed on

recovery

recovery relative to recovery

without bed

Influence of Vertical Permeability



BioturbationLenticular bedded

isolated lensesLenticular bedded

thick connected lensesWavy bedded

Shaly Sandstone

• Core through non-bioturbated interval would indicate good k in lenses

core

g g• Series flow indicates long-range permeability would be reduced to

permeability of shale k < 1μd• Bioturbation decreases k of lenses by 5-10X but preserves average k• Beneficial effect of bioturbation decreases with increasing sand:shale

ratio but amount of k decrease also decreases

Permeability ScalesPlug

Wireline- log

DST-Well Test

Lease-Reservoir

Establish role ofHeterogeneities& Fractures

Establish role ofHeterogeneities& F t

Full-DiameterCore

& Fractures



Conclusions• Drainage capillary pressure (Pc) can be modeled using

equations for threshold entry pressure (Pte) and Brooks-Corey λslopes.

• Capillary pressure (Pc) exhibits a log-log threshold entry (Pt ) k /φ t d d i bl B k Cpressure (Pte) versus kik/φi trend and variable Brooks-Corey

slopes. • Snwr ↑ with Snwi ↑ Land-type relation: 1/Snwr-1/Snwi = 0.55 •• Capillary pressure (Pc) is stress sensitive as expectedCapillary pressure (Pc) is stress sensitive as expected

–– threshold entry pressure is predictable from √K/threshold entry pressure is predictable from √K/ φφ at any at any confining pressureconfining pressure

•• Confining pressure decreases largest pores consistent with Confining pressure decreases largest pores consistent with g p g pg p g ppermeability decrease but has little influence on smaller pores permeability decrease but has little influence on smaller pores (pores largely protected by matrix)(pores largely protected by matrix)

•• Residual gas saturation increases with increasing initial gas Residual gas saturation increases with increasing initial gas saturationsaturation–– LandLand--type relation: (1/type relation: (1/SnwrSnwr))--(1/(1/SnwiSnwi) = 0.55 ) = 0.55

Conclusions• Multi-salinity measurements of Archie cementation exponent, m, have

been completed on 408 samples at various salinities for each sample– 20,000 ppm NaCl, 40,000 ppm, 80,000 ppm, and 200,000 ppm– Three times the number proposed

• Nearly all core exhibit some dependence of conductivity and cementation exponent on salinity

• The salinity dependence of m is weakly negatively correlated with porosity

• Using equations developed the Archie cementation exponent can be predicted for any given porosity and formation brine salinity

• Archie cementation exponent (m) decreases with decreasing porosity below approximately 6%

C b d l d i i l b d l it d l– Can be modeled- empirical or by a dual- porosity model



Conclusions• Analysis suggests that Sgc is scale- and bedding-architecture dependent in

cores and in the field. • Sgc is likely to be very low in cores with laminae and laminated reservoirs

( )) d l ( S 0 03 0 0 l d S 0 02 i(N//)) and low (e.g., Sgc < 0.03-0.07 at core scale and Sgc < 0.02 at reservoir scale) in massive-bedded sandstones of any permeability (Np)

• In cross-bedded lithologies exhibiting series network properties (N⊥), Sgcapproaches a constant reflecting the capillary pressure property differences and relative pore volumes among the beds in series. For these networks Sgccan range widely but can reach high values (e.g., Sgc < 0.6)

• Discontinuous series networks, representing lithologies exhibiting series t k ti b t f hi h th t i ti b d t lnetwork properties but for which the restrictive beds are not sample-

spanning (N⊥d), exhibit Sgc intermediate between Np and N⊥ networks.


Krygowski: Log Responses in Tight Shaly Gas Sands

Lithofacies and Petrophysical Lithofacies and Petrophysical P ti f M d Ti htP ti f M d Ti ht GGProperties of Mesaverde TightProperties of Mesaverde Tight--Gas Gas Sandstones in Western U.S. Basins:Sandstones in Western U.S. Basins:Log Responses in Tight Shaly Gas SandsLog Responses in Tight Shaly Gas Sands

Dan KrygowskiDan Krygowski


The geologic environmentThe geologic environmentComplicated lithology/mineralogyComplicated lithology/mineralogy

QuartzQuartzMixture of clays maybe diagenetic productsMixture of clays maybe diagenetic products

Quantities of interest

Mixture of clays, maybe diagenetic products Mixture of clays, maybe diagenetic products (Vcl/Vsh)

Low porosity, <15%Low porosity, <15% (Phi)FluidsFluids

Gas (water saturation, Sw < 1)Gas (water saturation, Sw < 1) (Sw)Relatively fresh waters Relatively fresh waters (Rw)


yy ( )High irreducible water saturation High irreducible water saturation (Swirr)

Permeability Permeability (k)Low, and of interestLow, and of interest



ExxonMobil Willow Ridge T63X-2G

A Mesaverde A Mesaverde exampleexample

2.65

Rio Blanco county, COPiceance Basin


Environmental effects on the logsEnvironmental effects on the logsComplicated lithology/mineralogyComplicated lithology/mineralogy

Presence of clayPresence of clay•• Gamma ray, SP: decreased response as compared to Gamma ray, SP: decreased response as compared to Ga a ay, S dec eased espo se as co pa ed toGa a ay, S dec eased espo se as co pa ed to

nearby shales.nearby shales.GR may also be affected by radioactive KGR may also be affected by radioactive K--feldspar.feldspar.

•• Porosity measurementsPorosity measurementsDensity porosity: slightly lowerDensity porosity: slightly lowerNeutron porosity: higherNeutron porosity: higherSonic porosity: higherSonic porosity: higher


•• Resistivity: lower, from additional clay conductivity.Resistivity: lower, from additional clay conductivity.May make water saturation calculations higher May make water saturation calculations higher than actual saturations.than actual saturations.



Environmental effects IIEnvironmental effects IIFluidsFluids

GasGas•• Gamma ray: no change. SP: decreased responseGamma ray: no change. SP: decreased response•• Density porosity: slightly higherDensity porosity: slightly higher•• Neutron porosity: lowerNeutron porosity: lower•• Sonic porosity: variableSonic porosity: variable

Relatively fresh waterRelatively fresh water•• Clay conductivity will be a larger percentage of the total Clay conductivity will be a larger percentage of the total

conductivity than in a salt water case.conductivity than in a salt water case.•• Resistivity decreased from equivalent clean case;Resistivity decreased from equivalent clean case;


y qy qShaly sand version of Archie needed?Shaly sand version of Archie needed?

High irreducible waterHigh irreducible water•• WaterWater--free production even with elevated water free production even with elevated water

saturations.saturations.

Environmental effects IIIEnvironmental effects IIIPermeabilityPermeability

Low, but of interestLow, but of interest•• Logs,even NMR logs, don’t measure permeability, but Logs,even NMR logs, don’t measure permeability, but ogs,e e ogs, do t easu e pe eab ty, butogs,e e ogs, do t easu e pe eab ty, but

we can infer permeability from log response. we can infer permeability from log response. •• Many equations; functions of porosity and irreducible Many equations; functions of porosity and irreducible

water saturation.water saturation.An example: Timur:An example: Timur:

•• We can get Swirr from BVWirr, irreducible bulk volume We can get Swirr from BVWirr, irreducible bulk volume water: BVW = Phi*Sw and BVWirr = Phi* Swirrwater: BVW = Phi*Sw and BVWirr = Phi* Swirr

⎟⎟⎠

⎞⎜⎜⎝

⎛∗= 2

662500

irrT

SwPhiK


water: BVW = Phi Sw, and BVWirr = Phi Swirrwater: BVW = Phi Sw, and BVWirr = Phi Swirr

and BVW can give us some indication of fluids and BVW can give us some indication of fluids that will be produced (water vs no water).that will be produced (water vs no water).



Quantities, parameters of interestQuantities, parameters of interestClay/shale volumeClay/shale volume

Density/neutron: problematic because of gas Density/neutron: problematic because of gas effects on the neutron.effects on the neutron.

•• In general, neutron porosity has issues in the Rockies.In general, neutron porosity has issues in the Rockies.

olum

e, V

sh

SP: hydrocarbon SP: hydrocarbon effects will make effects will make Vsh too high.Vsh too high.Gamma ray: Gamma ray: probably the best. probably the best.

AAPG ACE 2009: Denver Colorado 7Radioactivity Index, IRAGamma Ray Index, IGR

Shal

e vp yp y

Use linear unless Use linear unless other data indicates other data indicates otherwise.otherwise.

Vsh may be needed for the following quantities...

BakerAtlas, 1984

More quantities of interestMore quantities of interestPorosity, PhiPorosity, Phi

Need matrix and fluid parametersNeed matrix and fluid parameters•• Variable matrix parameters are not uncommon.Variable matrix parameters are not uncommon.a ab e at pa a ete s a e ot u co oa ab e at pa a ete s a e ot u co o

May need shale/clay parameters: Vsh, shale May need shale/clay parameters: Vsh, shale values for specific measurements: density, values for specific measurements: density,

tt

RHOflRHOmaRHOBRHOmaPHID

−−

=

DTDTmaDT

DTmaDTflDTmaDTPHIS −

=−−

= *32 or


neutron, …neutron, …•• Effective porosity from total porosity, Vsh, and shale Effective porosity from total porosity, Vsh, and shale

response.response.SHeff PHIDVshPHIDPHID ∗−=



Porosity in a gas zonePorosity in a gas zoneSingle porosity measurementSingle porosity measurement

Can the matrix and fluid parameters in the Can the matrix and fluid parameters in the volume of investigation be sufficiently estimated volume of investigation be sufficiently estimated g yg yto produce a reasonable porosity?to produce a reasonable porosity?

•• Most porosity measurements are in the flushed zone.Most porosity measurements are in the flushed zone.

Porosity measurement combinations: Porosity measurement combinations: density and neutrondensity and neutron

If the neutron is good, this is actually a good If the neutron is good, this is actually a good ti t f h d bti t f h d b t d l tt d l t


estimate of hydrocarbonestimate of hydrocarbon--corrected crossplot corrected crossplot porosity.porosity.

21

22

2 ⎟⎟⎠

⎞⎜⎜⎝

⎛ +=

PHINePHIDePHIE

More quantities, for saturationMore quantities, for saturationWater saturation, SwWater saturation, Sw

Water resistivity, RwWater resistivity, Rw•• Produced waters yield Rw values that are much too Produced waters yield Rw values that are much too

fresh (water of condensation in the gas).fresh (water of condensation in the gas).•• NOT SP! NOT SP! Rwa vs GR

75

100

125

150

GR

RwbGRshale•• Pickett plot or Rwa, Pickett plot or Rwa,

apparent water resistivity apparent water resistivity

Archie parameters, a, Archie parameters, a, m (variable), nm (variable), n•• Local knowledge; PickettLocal knowledge; Pickett

AAPG ACE 2009: Denver Colorado 10If Rwf = Rwb, use Archie.

0

25

50

0.1 1 10 100

Rwa

Rw, Rwf

data

GRclean

•• Local knowledge; Pickett Local knowledge; Pickett plotplot

Which form of Archie’s Which form of Archie’s equation?equation?•• Vsh & Rsh; or Rwf & Vsh & Rsh; or Rwf &

RwbRwb



Rwa in the MesaverdeRwa in the MesaverdeRwb


Rw, Rwf

Another saturation parameter methodAnother saturation parameter method

“Super Pickett” plot“Super Pickett” plotGetting a number of parameters.Getting a number of parameters.BVWirr canBVWirr can

Pickett plot

0.1

1

rosi

ty data

reas

ing

Sw

Rw BVWirrincreasing BVW

Slope = f(saturation exponent,n)BVWirr can BVWirr can also be also be estimated estimated from a log from a log plot.plot.


0.011 10 100 1000

Resistivity

Po

Sw = 1

decr

Slope = -1/cementation exponent, m



Pickett with Mesaverde dataPickett with Mesaverde data

Sw = 1 0 6 0 4 0 2

BVW = 0.10.05

0.04BVWirr = 0.026

From slope,saturation exponent, n = 2.0

Rw = 0.064

Sw = 1 0.6 0.4 0.2

From slope,cementation exponent, m = 1.85


Which saturation equation to use?Which saturation equation to use?The most commonly used in the Rockies:The most commonly used in the Rockies:

ArchieArchie21⎞⎛ ∗Rwa

In conductivity spaceIn conductivity space(Ct = 1000/Rt):(Ct = 1000/Rt):

Dual WaterDual Water

2⎟⎠

⎞⎜⎝

⎛∗

∗=

RtPhiRwaSw m CwPhiSwaCt mn ∗∗∗=

⎤⎡ ⎞⎛

SwSw = [a number of versions are published…]= [a number of versions are published…]


⎥⎦

⎤⎢⎣

⎡∗+∗⎟

⎠

⎞⎜⎝

⎛ −∗∗= CwbSwSwbCwf

SwSwbPhiSwCt mn 1



Mesaverde Mesaverde data againdata again

What about permeability?What about permeability?Timur (and other equations) requires Swirr.Timur (and other equations) requires Swirr.

•• Swirr is a proxy for surface area.Swirr is a proxy for surface area.We can get Swirr from BVWirr:We can get Swirr from BVWirr:

But the permeability numbers are suspect (at But the permeability numbers are suspect (at best).best).

•• Core data is needed to calibrate the permeability Core data is needed to calibrate the permeability calculation, calibration being done by modifying the calculation, calibration being done by modifying the porosity and saturation exponents.porosity and saturation exponents.

PHIBVWirrSwirr /=


porosity and saturation exponents.porosity and saturation exponents.NMR logs can provide permeabilityNMR logs can provide permeability

•• They measure both Phi and BVWirr.They measure both Phi and BVWirr.•• But they still need calibration to core for quantitative But they still need calibration to core for quantitative

values.values.



…and bulk volume water, BVW……and bulk volume water, BVW…If Sw < 1, and BVW is a constant, the zone If Sw < 1, and BVW is a constant, the zone has a good chance of producing waterhas a good chance of producing water--free.free.

But we can’t determine the production volumesBut we can’t determine the production volumesBut we can t determine the production volumes.But we can t determine the production volumes.If BVW > 0.05, there’s a good chance that the If BVW > 0.05, there’s a good chance that the well will produce no fluids at all.well will produce no fluids at all.

•• Pore throats are blocked by water.Pore throats are blocked by water.


Mesaverde Mesaverde with with permeability permeability and BVWand BVW



ConclusionsConclusionsThe combination of gas, shaly formations, The combination of gas, shaly formations, and low porosity has adverse affects on all and low porosity has adverse affects on all the logging measurements.the logging measurements.the logging measurements.the logging measurements.

Some of the effects counteract each other; i.e., Some of the effects counteract each other; i.e., gas and clays on neutron porosity.gas and clays on neutron porosity.Generally, the difference between wet zones and Generally, the difference between wet zones and pay is more subtle.pay is more subtle.


So, what specifically have we learned about So, what specifically have we learned about the Mesaverde in the Rockies?the Mesaverde in the Rockies?

The story continues…The story continues…


Whittaker: Standard Log Analysis

Lithofacies and Petrophysical Lithofacies and Petrophysical Properties of Mesaverde TightProperties of Mesaverde Tight GasGasProperties of Mesaverde TightProperties of Mesaverde Tight--Gas Gas Sandstones in Western U.S. Basins:Sandstones in Western U.S. Basins:Standard AnalysisStandard Analysis

Stefani WhittakerStefani Whittaker


OUTLINEOUTLINE

DATA PREPARATION

Gather Data and Initial Clean up Calc. In situ Core DataImport corrected core data, rock type numbers, and point count numbersShifting: Core data, point count data and rock type data


type data Pick tops and zonesSetting up zone parameters



CALCULATION:

Calculate VshCalculate VshTotal and Effective Porosities Calculate SwLook at a Pickett Plot Calculate SWICalculate perm


Calculate perm

Gathering WellGathering Well--Log DataLog Data

Required Curves Required Curves

Depth Matching Depth Matching p gp g

Merging Multiple RunsMerging Multiple Runs

Tool PickTool Pick--up up

Neutron Matrix ConversionNeutron Matrix Conversion


NormalizationNormalization



Calculating Calculating InIn--situsitu Core DataCore DataKlinkenbergKlinkenberg CorrectedCorrected

008.0−=CPHICPHIinsitu

Porosity

Permeability

6.0)(log341.1log −= routineinsitu kK


*Note: Alan Byrnes equation from The Mountain Geologist; Volume 34; Number 1; “Reservoir Characteristics of Low-Permeability Sandstones in the Rocky Mountains”; pg. 42. There is a mistype in the publication, the above equation is the CORRECT equation.

Importing DataImporting Data1)1) In Situ Core Data In Situ Core Data

●● Conventional Core DataConventional Core Data●● KGS analyzed Core DataKGS analyzed Core Data (Appended _KGS)(Appended _KGS)

2)2) Rock Type DataRock Type Data•• Core description 5 digit rock type codeCore description 5 digit rock type code•• 5 digit code can be compared to GR5 digit code can be compared to GR


3)3) Point Count DataPoint Count Data•• Thin Section Point Count Data Thin Section Point Count Data •• The total radiation term (VRAD_TS) can The total radiation term (VRAD_TS) can

be compared to the Vsh curve in the logs.be compared to the Vsh curve in the logs.



VANHCMT_TSVANHCMT_TS Volume of anhydrite in thin sectionVolume of anhydrite in thin section

VCCMT_TSVCCMT_TS Volume of clay cement in thin sectionVolume of clay cement in thin section

VCO3CMT_TSVCO3CMT_TS Volume of carbonate cement in thin sectionVolume of carbonate cement in thin section

VKSP_TSVKSP_TS Volume of Potassium Feldspar in thin sectionVolume of Potassium Feldspar in thin section

VKVRF_TSVKVRF_TS Volume of Potassium rich volcanic rock fragments in thin sectionVolume of Potassium rich volcanic rock fragments in thin section

VOSRF_TSVOSRF_TS Volume of of other sedimenary rock fragments in thin sectionVolume of of other sedimenary rock fragments in thin section

VOVRF_TSVOVRF_TS Volume of other volcanic rock fragments in thin sectionVolume of other volcanic rock fragments in thin section

VPLAG_TSVPLAG_TS Volume of Plagioclase Feldspars in thin sectionVolume of Plagioclase Feldspars in thin section

VQTZ TSVQTZ TS Volume of quartz in thin sectionVolume of quartz in thin sectionVQTZ_TSVQTZ_TS Volume of quartz in thin sectionVolume of quartz in thin section

VQTZCMT_TSVQTZCMT_TS Volume of quartz cement in thin sectionVolume of quartz cement in thin section

VRAD_TSVRAD_TS Volume of Radioactive Elements in thin section Volume of Radioactive Elements in thin section (VRAD_TS = VKSP_TS + VKVRF_TS + VSSRF_TS + VCCMT_TS + VOVRF_TS)(VRAD_TS = VKSP_TS + VKVRF_TS + VSSRF_TS + VCCMT_TS + VOVRF_TS)

VSSRF_TSVSSRF_TS Volume of Shaley sedimentary rock fragments in thin sectionVolume of Shaley sedimentary rock fragments in thin section

VVISPOR_TSVVISPOR_TS Volume of Visible Porosity in thin sectionVolume of Visible Porosity in thin section

Depth Shifting Core DataDepth Shifting Core Data

Rock Type Number was compared to the GR.Rock Type Number was compared to the GR.

Data Shifted together:Data Shifted together:•• Conventional Core DataConventional Core Data•• KGS analyzed Core DataKGS analyzed Core Data•• Point Count DataPoint Count Data•• Rock Type DataRock Type Data


Rock Type DataRock Type Data



Picking Tops and ZonesPicking Tops and Zones

11 PIPI DwightsDwights scout tickets for formation topsscout tickets for formation tops1.1. PI PI DwightsDwights scout tickets for formation tops.scout tickets for formation tops.

2.2. Zones were chosen based on changes in Zones were chosen based on changes in petrophysicalpetrophysical properties to “tighten the properties to “tighten the log/core correlation log/core correlation

•• GRGR


•• GRGR•• PorosityPorosity•• InductionInduction

Standard Discovery Group Standard Discovery Group Shaly Sand ProcessShaly Sand Process

1.1. Set up ParametersSet up Parameters22 Calculate VshaleCalculate Vshale2.2. Calculate VshaleCalculate Vshale3.3. Calculate Porosity Calculate Porosity (Total, Effective, Cross(Total, Effective, Cross--Plot)Plot)

4.4. Calculate Water SaturationCalculate Water Saturation5.5. Calculate Bulk Volume Water andCalculate Bulk Volume Water and

Bulk Volume Water IrreducibleBulk Volume Water Irreducible


and Calculate Irreducible Water Saturationand Calculate Irreducible Water Saturation6.6. Calculate PermeabilityCalculate Permeability



Setting up zone ParametersSetting up zone ParametersDeep ResistivityDeep Resistivity Rt = RdeepRt = Rdeep

Rho Matrix Rho Matrix From Header DataFrom Header Data

Neutron MatrixNeutron Matrix From Header DataFrom Header DataNeutron Matrix Neutron Matrix From Header Data From Header Data

Vshale Model Vshale Model Linear using GRLinear using GR

Water Sat. Model Water Sat. Model Archie’s (m=1.85, n=2, a=1)Archie’s (m=1.85, n=2, a=1)

BVW Model BVW Model Effective PorosityEffective Porosity


Permeability ModelPermeability Model Timur ModelTimur Model

Parameters for Permeability were varied by zone:•Permeability Porosity Exponent [KPHIEXP] (Ranged from 5.0 - 9.25)

•Permeability Irreducible Water Saturation Exponent [KSWIEXP] (Ranged from 1.5 - 2.0)

Calculate VshCalculate Vsh

Used the GR with the Linear method to calculate Vsh. Used the GR with the Linear method to calculate Vsh.

Rocky Mountain Region Suggestions:Rocky Mountain Region Suggestions:

cleansh

cleansh GRGR

GRGRV

−

−= log


y g ggy g ggGR_CLEAN = 10GR_CLEAN = 10--15 API15 APIGR_SHALE = 90GR_SHALE = 90--100 API100 API

(Will vary from well to well)(Will vary from well to well)



Total PorosityTotal Porosity

Total Porosity Total Porosity PHIN = Converted from LS units toPHIN = Converted from LS units toPHIN = Converted from LS units to PHIN = Converted from LS units to desired output desired output lithologylithology units.units.

PHID = PHID = RHOFLRHOMARHOBRHOMA

−− (Wyllie Time (Wyllie Time

Average Equation)Average Equation)


PHIS = PHIS = DTMADTFDTMAt

−−Δ log

•Take RHOB and Neutron Φ and cross plot them to get a PHIDN

Cross-Plot Porosities

PROS:

-Corrects for grain density

-Eliminates most of the gas effect

CONS:

-Requires a good NPHI log



Effective PorosityEffective Porosity

)*( PHINSHVPHINPHINE sh−=

)*( PHIDSHVPHIDPHIDE sh−=

)*( PHISSHVPHISPHISE sh−=


)*( PHIDNSHVPHIDNPHIDNE sh−=

(Diminish the effect of Shale)(Diminish the effect of Shale)

Total Φ Diminishes Shale Volume Diminishes1. Grain Density Differences2. Gas Effect3. Shale Volume



Calculate SwCalculate SwArchie’s Water Saturation equationArchie’s Water Saturation equation

a=1; n=2; m=1.85 (Rocky Mountain Suggestion)a=1; n=2; m=1.85 (Rocky Mountain Suggestion)RwRw = Zoned (Pickett Plot or = Zoned (Pickett Plot or RwaRwa plot)plot)Used Neutron/Density Used Neutron/Density crossplotcrossplot Effective PorosityEffective PorosityRtRt = Deep Resistivity= Deep Resistivity

n waRSw =


n

tmR

Swφ

BVW, BVWI and SWIBVW, BVWI and SWITwo ways to find BVW, BVWI, and SWITwo ways to find BVW, BVWI, and SWI

1) Calculate and visual estimation1) Calculate and visual estimation2) Graphically using Pickett Plot2) Graphically using Pickett Plot

Calculate:Calculate:

we SPHIEBVW *=wT SPHIXBVW *=


Then look at a consistently flat part on the BVW and Then look at a consistently flat part on the BVW and visually pick the BVWIvisually pick the BVWI

PHIBVWISWI /=



Pickett PlotPickett Plot100% Water Sat. when a=1

Iso BVW lines

BVWI



Calculate PermeabilityCalculate Permeability

Used the Timur Model for permeabilityUsed the Timur Model for permeability

62500=coefK

25.90.5~ −KPHIEXPK

0.25.1~ −KSWIEXPK

(Determined by zone)

(Determined by zone)


KSWIEXP

KPHIEXP

coef SWIPHIXKK =log

Piceance BasinPiceance Basin

Vshale Φ, m&nΦ, SWIKexp.

Left to right more error introducedLeft to right more error introducedError introduced =



Green River BasinGreen River Basin

Washakie BasinWashakie Basin



Uinta BasinUinta Basin

Wind River BasinWind River Basin


Cluff: Advanced Log Models

Lithofacies and Petrophysical Lithofacies and Petrophysical Properties of Mesaverde TightProperties of Mesaverde Tight GasGasProperties of Mesaverde TightProperties of Mesaverde Tight--Gas Gas Sandstones in Western U.S. Basins:Sandstones in Western U.S. Basins:Advanced Log AnalysisAdvanced Log Analysis

Bob CluffBob CluffThe Discovery Group IncThe Discovery Group Inc


The Discovery Group Inc.The Discovery Group Inc.2009 AAPG Annual Convention Short course #12009 AAPG Annual Convention Short course #1

6 June 2009, Denver, Colorado6 June 2009, Denver, Colorado

OutlineOutlinerock typingrock typingvariable m model for Swvariable m model for Sw

as an alternative to obtuse shaly sand modelsas an alternative to obtuse shaly sand modelsas an alternative to obtuse shaly sand modelsas an alternative to obtuse shaly sand modelspermeability modelingpermeability modeling




Advanced rock typingAdvanced rock typingmost rock typing methods follow some form of most rock typing methods follow some form of φφ--K K separation or BVW separationseparation or BVW separation

Winland R35 isoWinland R35 iso--lineslinesK/K/φφ ratiosratiosBVW classesBVW classes

or, some kind of statistical relationship with logs is or, some kind of statistical relationship with logs is soughtsought

single variate comparsions (e.g. GR vs grain size)single variate comparsions (e.g. GR vs grain size)multivariate comparisons, cluster analysis, etc.multivariate comparisons, cluster analysis, etc.


neural networks (a fancy form of multivariate nonneural networks (a fancy form of multivariate non--linear linear regression)regression)

Winland equationWinland equationDeveloped by Amoco in 1970’sDeveloped by Amoco in 1970’sEmpirically derived eqn from a large Pc dataset, Empirically derived eqn from a large Pc dataset, Weyburn field in CanadaWeyburn field in CanadaEqn published by Kolodzie, 1980 (SPE 9382) Eqn published by Kolodzie, 1980 (SPE 9382) Rock types defined by “equiRock types defined by “equi--pore throat size” pore throat size” classes, or “port” sizes, as determined from Pc at classes, or “port” sizes, as determined from Pc at 35% Snw35% Snw

macroports = 2macroports = 2--10 10 μμmmmesoports = 0.5 mesoports = 0.5 –– 2 2 μμmm


pp μμmicroports = 0.1 microports = 0.1 –– 0.5 0.5 μμmmnanoports < 0.1 nanoports < 0.1 μμmm

implicit is pore throat sizes control hydrocarbon implicit is pore throat sizes control hydrocarbon entry and relate to pay qualityentry and relate to pay quality



100

1000

D)

R35

Winland R35 “port” size classesWinland R35 “port” size classes

“macroports”

log R35 = 0.732 + 0.588 log Kair log R35 = 0.732 + 0.588 log Kair –– 0.864 log 0.864 log φ φ ((%)%)

0.001

0.01

0.1

1

10

nken

berg

gas

per

mea

bilit

y (M

D

2

0.5

0.1

0.02

“microports”

“nanoport”


0.000001

0.00001

0.0001

0.0 5.0 10.0 15.0 20.0 25.0

in-situ porosity (%)

in-s

itu K

lin

Note: essentially all Kmv TGSfall into the nanoport rock type

100

1000

(MD

) K/phi

K/K/φ φ ratio isoratio iso--lineslinesK/phi ratio = Ka (mD) / φ (v/v)

0.001

0.01

0.1

1

10

Klin

kenb

erg

gas

perm

eabi

lity

( 50

5

0.5

0.05

0.005


0.000001

0.00001

0.0001

0.0 5.0 10.0 15.0 20.0 25.0

in-situ porosity (%)

in-s

itu

Note: most smpls are at K/f < 0.5and would fall into 3 or 4 classes,but without natural breaks



K/phi methodsK/phi methodsyou can compute K/phi ratio from ambient or inyou can compute K/phi ratio from ambient or in--situ situ core data, or from log K and phicore data, or from log K and phi

divide it into classes that make sense for your areadivide it into classes that make sense for your areano natural divisions in the overall databaseno natural divisions in the overall database

compute Winland R35 from standard eqn or cook compute Winland R35 from standard eqn or cook your own eqn from our dataset!your own eqn from our dataset!

we have NOT done this for youwe have NOT done this for youLOTS of ways to slice and dice this large a databaseLOTS of ways to slice and dice this large a database

basic Winland classes have limited utility in very basic Winland classes have limited utility in very


tight rocks like these, almost everything falls into tight rocks like these, almost everything falls into the “nanoport” size rangethe “nanoport” size range

Rock types from logsRock types from logswe have digital rock types from core we have digital rock types from core description depth shifted to log datadescription depth shifted to log dataseems like we should be able to pull rockseems like we should be able to pull rockseems like we should be able to pull rock seems like we should be able to pull rock types out of the log data by xtypes out of the log data by x--plots or plots or statistical analysisstatistical analysisWell, maybe its not so easy.........Well, maybe its not so easy.........




Digital core database @ 0.5 ft resolutionDigital core database @ 0.5 ft resolution

GR log plot vs rock #GR log plot vs rock #

GR to rock # correlation is outstanding!GR to rock # correlation is outstanding!



GR vs Rock numberGR vs Rock number

but over the entire database, therock type classes broadly overlap

Why is that?Why is that?GR logs are not normalizedGR logs are not normalized

it looks good on a single well basis, but gets it looks good on a single well basis, but gets smeared out over multiple cores/wellssmeared out over multiple cores/wellsppuncorrected environmental effectsuncorrected environmental effectsall vendors GR tools are not alikeall vendors GR tools are not alike

the 13000 rock class will always be a the 13000 rock class will always be a problem, by nature of the definition they problem, by nature of the definition they span a broad range of Vshspan a broad range of Vsh


p gp gonly the higher rock classes (1only the higher rock classes (1stst 2 or 3 2 or 3 digits) are likely to fall out in the best of digits) are likely to fall out in the best of casescases



ILD vs. GR xplot colored by major rock #ILD vs. GR xplot colored by major rock #

11000 to 12999’s separate cleanly from 15000’s, butthe 13000’s overlap all

NPHINPHI--RHOB by major rock #RHOB by major rock #

again the 15000’s splitcleanly from 12000’s,while 13000’s overlapthe entire field



DT DT -- RHOB colored by major rock #RHOB colored by major rock #

nothing separates on this,because DT and RHOBare too similar in their lithology response

Rock typing summaryRock typing summarythere is a lot of data here, we didn’t push the there is a lot of data here, we didn’t push the boundaries of what could be done by any boundaries of what could be done by any meansmeansmeansmeansBUT, from our analysis, the results do not BUT, from our analysis, the results do not look promisinglook promisingvery, very difficult to pull out subtle rock type very, very difficult to pull out subtle rock type signatures from a limited suite of open hole signatures from a limited suite of open hole measurements if the base lithology does notmeasurements if the base lithology does not


measurements if the base lithology does not measurements if the base lithology does not change muchchange muchonly grain size comes out cleanly, but with a only grain size comes out cleanly, but with a broad overlap between classesbroad overlap between classes



Saturation modelSaturation modelbasic model assumes Archie with TGS basic model assumes Archie with TGS average m, n valuesaverage m, n valuesShaly sand models (e g Dual Water) allShaly sand models (e g Dual Water) allShaly sand models (e.g. Dual Water) all Shaly sand models (e.g. Dual Water) all yield similar results because fm. waters are yield similar results because fm. waters are saline and shales are not highly conductivesaline and shales are not highly conductivecore data suggests m varies as a function of core data suggests m varies as a function of both porosity and average salinityboth porosity and average salinity


1000

When F and When F and φφ are plotted logare plotted log--loglog

m= 3m= 2

10

100

Fm= 1


10.01 0.1 1

φlog F = -m log φ



Salinity dependence of “m”Salinity dependence of “m”

tested plugs with 20K, 40K, 80K, and 200K ppm brinestested plugs with 20K, 40K, 80K, and 200K ppm brinesNearly all cores exhibit some salinity dependenceNearly all cores exhibit some salinity dependence

1.0

2 2

2.3

,

0.3

0.4

0.5

0.6

0.7

0.8

0.9

e C

ondu

ctiv

ity (m

ho/m

)

1.4

1.5

1.6

1.7

1.8

1.9

2.0

2.1

2.2

chie

Cem

enta

tion

Expo

nent

(m, A

=1)

n=335


0.0

0.1

0.2

0 2 4 6 8 10 12 14 16 18 20 22

Brine Conductivity (mho/m)

Cor

e

1.0

1.1

1.2

1.3

0.01 0.1 1

Brine Resistivity (ohm-m)In

situ

Arc

All data, all salinities All data, all salinities

2.20

2.40

m, a

=1)

1.20

1.40

1.60

1.80

2.00

Cem

enta

iton

Expo

nent

(m

200K


0.80

1.00

0 2 4 6 8 10 12 14 16 18 20 22


Arc

hie

C

80K

40K

20K



Salinity dependence of “m”Salinity dependence of “m”

m = a log m = a log φ φ + b+ bintercept b drops with intercept b drops with decreasing salinitydecreasing salinity

20K ppm

y = 0.2267Ln(x) + 2.2979

R2 = 0.6619

1.00

1.50

2.00

2.50

Axis Title

Series1

Log. (Series1) decreasing salinitydecreasing salinityslope is ~ constantslope is ~ constant0.00

0.50

0.000 0.050 0.100 0.150 0.200 0.250

insitu porosity (%)

40K ppm

y = 0.2328Ln(x) + 2.409

R2 = 0.6547

0.00

0.50

1.00

1.50

2.00

2.50

3.00

Axis Title

Series1

Log. (Series1)

80K ppm

y = 0.2149Ln(x) + 2.43542

3.00

200K ppm

y = 0.1621Ln(x) + 2.3222

3.00


0.000 0.050 0.100 0.150 0.200 0.250

insitu porosity (%)

R2 = 0.5132

0.00

0.50

1.00

1.50

2.00

2.50

0.000 0.050 0.100 0.150 0.200 0.250

insitu porosity (%)

Axis Title

Series1

Log. (Series1)

y 0.1621Ln(x) + 2.3222

R2 = 0.3633

0.00

0.50

1.00

1.50

2.00

2.50

0.000 0.050 0.100 0.150 0.200 0.250

insitu porosity (%)

Axis Title

Series1

Log. (Series1)

Simple procedure to compute SwSimple procedure to compute Sw

determine Rw @ Tf conventionallydetermine Rw @ Tf conventionallyPickett plots Pickett plots –– focus on the lower porosity, wetter focus on the lower porosity, wetter sandstonessandstonesproduced watersproduced watersyour best guess.......your best guess.......

convert Rw to 75convert Rw to 75°°F by chart lookup or Arps F by chart lookup or Arps equationequation




Pickett Plot examplePickett Plot example

Rw = 0.306

pick m at low porosityend, where BVWirr ~ BVW

Williams PA 424Williams PA 424--3434Piceance basinPiceance basinKmv above “top gas”Kmv above “top gas”

Pickett plot Rw 0.306 ohmm @ 160Pickett plot Rw 0.306 ohmm @ 160°°F = 0.7 @ 75F = 0.7 @ 75°°F (9K ppm) F (9K ppm)



Our new procedureOur new procedurecompute m at 40K ppm from RMA regression:compute m at 40K ppm from RMA regression:

m40k = 0.676 log m40k = 0.676 log φ φ + 1.22+ 1.22e.g. for 10% e.g. for 10% φ : φ : m = 0.676 + 1.22 = 1.896m = 0.676 + 1.22 = 1.896

correct m for salinit effect bcorrect m for salinit effect bcorrect m for salinity effect bycorrect m for salinity effect bym = m40k + ((0.0118 m = m40k + ((0.0118 φφ –– 0.355) * (log Rw + 0.758))0.355) * (log Rw + 0.758))

e.g. for 10% e.g. for 10% φφ, Rw = 0.7 @ 75, Rw = 0.7 @ 75°°FFm = 1.896 + ((0.0118 * 10 m = 1.896 + ((0.0118 * 10 –– 0.355) * (log 0.7 + 0.758))0.355) * (log 0.7 + 0.758))m = 1.896 + (m = 1.896 + (--0.237 * 0.603) = 1.7530.237 * 0.603) = 1.753

cap m at 1.95 (~12% porosity)cap m at 1.95 (~12% porosity)this corrects for variation in both porosity and fmthis corrects for variation in both porosity and fm


this corrects for variation in both porosity and fm this corrects for variation in both porosity and fm salinity spacesalinity space

Practical impactPractical impactNominally, most of us use an m close to 2, Nominally, most of us use an m close to 2, but usually slightly less, for tight gas sand but usually slightly less, for tight gas sand evaluations (evaluations (e.g.e.g. 1.85, 1.90)1.85, 1.90)Variable m that DECREASES with Variable m that DECREASES with decreasing porosity leads to lower Sw’sdecreasing porosity leads to lower Sw’sTherefore, there is more gas in the tight Therefore, there is more gas in the tight rocks than we thought.rocks than we thought.Above 10% porosity there is very little Above 10% porosity there is very little differencedifference


differencedifference



Example: Low porosity, wet zoneExample: Low porosity, wet zone

Moderate porosity, wetModerate porosity, wet



“High” porosity gas zone“High” porosity gas zone

m is HIGHER than base case, so Sw is higher!

20Kppm example, Natural Buttes20Kppm example, Natural Buttes

improvement in HCPV in shoulders



30K ppm example, Wamsutter30K ppm example, Wamsutter

no change

Sw summarySw summary335 Kmv samples run at multiple salinities335 Kmv samples run at multiple salinitiesArchie porosity exponent m varies withArchie porosity exponent m varies with

porosityporosity mm ↓ as porosity ↓↓ as porosity ↓porosity porosity m m ↓ as porosity ↓↓ as porosity ↓salinitysalinity m m ↓ as salinity ↓↓ as salinity ↓

behavior is consistent with increasing behavior is consistent with increasing electrical efficiency with decreasing porosity, electrical efficiency with decreasing porosity, whatever the pore scale architecturewhatever the pore scale architecture

very likely that the surface conductivity is highlyvery likely that the surface conductivity is highly


very likely that the surface conductivity is highly very likely that the surface conductivity is highly connected with low effective mconnected with low effective mporepore--pore throat conductivity is Archie with m pore throat conductivity is Archie with m close to 2close to 2



Capillary tube model for mCapillary tube model for m

m 1.0

> 1

~2

> 2m = 1


Herrick & Kennedy, 1993, SPWLA Paper HH

E0 vs porosity, 40K ppm dataE0 vs porosity, 40K ppm data

TableCurve 2D v5.01



variable m Archie model can be implemented with a variable m Archie model can be implemented with a simple equation relating m to porosity and formation simple equation relating m to porosity and formation water salinitywater salinitym is constant above 12% porosity at 1 95m is constant above 12% porosity at 1 95m is constant above ~12% porosity at 1.95m is constant above ~12% porosity at 1.95lowering m at 5lowering m at 5--12% 12% φ φ increases GIPincreases GIPsee no impact below ~5% porosity see no impact below ~5% porosity

BVWBVWirrirr is typically 3is typically 3--5%5%no longer calculate Sw’s >> 1no longer calculate Sw’s >> 1Sw = 1 at low Sw = 1 at low φφ validates Rwvalidates Rw

much simpler than Dual Water or Wmuch simpler than Dual Water or W--S formulationsS formulations


much simpler than Dual Water or Wmuch simpler than Dual Water or W--S formulations S formulations for TGS, easier to implement, and it gets you the for TGS, easier to implement, and it gets you the same answersame answer

PermeabilityPermeabilitypermeability has historically been a problem permeability has historically been a problem to estimate from log datato estimate from log datadynamic property that we are trying todynamic property that we are trying todynamic property that we are trying to dynamic property that we are trying to correlate with static propertiescorrelate with static properties

problem is there are no 1:1 functional problem is there are no 1:1 functional relationships between any of the static relationships between any of the static properties, like porosity, and permeability.properties, like porosity, and permeability.

so, we fudge....so, we fudge....


gg



Permeability from logsPermeability from logsPorosityPorosity--permeability crosspermeability cross--plotsplots

regression equations developed for each basin regression equations developed for each basin and presented previouslyand presented previouslyp p yp p ywith an accurate log porosity, you can predict K with an accurate log porosity, you can predict K within a SE of about 4X to 5Xwithin a SE of about 4X to 5Xif you add information such as grain size or rock if you add information such as grain size or rock type, you can do even bettertype, you can do even betteronly a fraction of what is possible to do has been only a fraction of what is possible to do has been done but basic eqn’s by basin are presented indone but basic eqn’s by basin are presented in


done, but basic eqn s by basin are presented in done, but basic eqn s by basin are presented in the project data storethe project data store

10

100

1000

si, m

D)

0 00001

0.0001

0.001

0.01

0.1

1

10

berg

Per

mea

bilit

y (4

,000

ps

Green RiverPiceancePowder RiverUintahWashakieWind River


0.0000001

0.000001

0.00001


Klin

kenb logK=0.3Phi-3.7

logK=0.3Phi-5.7



Kozeny & TimurKozeny & Timur--type eqn’stype eqn’sKozeny equationKozeny equation

K = A * K = A * φφ33 / S/ S22, , where S = surface area/bulk volumewhere S = surface area/bulk volume

Timur eqn (and its derivatives) are of this general Timur eqn (and its derivatives) are of this general form, but use Swi as a proxy for the internal surface form, but use Swi as a proxy for the internal surface area termarea term

K = 0.136 * K = 0.136 * φφ4.44.4 / Swi/ Swi22 (original Timur eqn)(original Timur eqn)K = 62,500 * K = 62,500 * φφ66 / Swi/ Swi22 (Schlumberger eqn)(Schlumberger eqn)K = A * K = A * φφBB / Swi/ SwiCC (general form)(general form)

We treat A, B, C as local variables and fit parameters by trial We treat A, B, C as local variables and fit parameters by trial


p yp yand error or using a multivariate solver (e.g. Excel Solver)and error or using a multivariate solver (e.g. Excel Solver)note:note: NMR eqn’s (e.g. Coates & SDR or T2GM) are basically NMR eqn’s (e.g. Coates & SDR or T2GM) are basically the general Timur eqn, but use Swi and the general Timur eqn, but use Swi and φ φ from NMR instead from NMR instead of indirect estimatesof indirect estimates




Thank you!Thank you!Q&A period (Q&A period (if time availableif time available))

Visit our project website portalsVisit our project website portals::http://www.kgs.ku.edu/mesaverdehttp://www.kgs.ku.edu/mesaverde

ororhttp://www.discoveryhttp://www.discovery--group.com/projects_doe.htmgroup.com/projects_doe.htm



lithofacies and petrophysical properties of mesaverde tight-gas

Documents